You are on page 1of 39

ConnorsResearchTradingStrategySeries

ShortSellingStocks
withConnorsRSI
By
ConnorsResearch,LLC
LaurenceConnors
CesarAlvarez
MattRadtke

P a g e |2

Copyright 2013, Connors Research, LLC.


ALL RIGHTS RESERVED. No part of this publication may be reproduced, stored in a
retrieval system, or transmitted, in any form or by any means, electronic, mechanical,
photocopying, recording, or otherwise, without the prior written permission of the
publisher and the author.
This publication is designed to provide accurate and authoritative information in regard
to the subject matter covered. It is sold with the understanding that the author and the
publisher are not engaged in rendering legal, accounting, or other professional service.
Authorization to photocopy items for internal or personal use, or in the internal or
personal use of specific clients, is granted by Connors Research, LLC, provided that the
U.S. $7.00 per page fee is paid directly to Connors Research, LLC, 1-973-494-7333.
ISBN 978-0-9886931-5-9
Printed in the United States of America.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |3

Disclaimer
By distributing this publication, Connors Research, LLC, Laurence A. Connors, Cesar Alvarez, and Matt
Radtke (collectively referred to as Company") are neither providing investment advisory services nor
acting as registered investment advisors or broker-dealers; they also do not purport to tell or suggest
which securities or currencies customers should buy or sell for themselves. The analysts and employees
or affiliates of Company may hold positions in the stocks, currencies or industries discussed here. You
understand and acknowledge that there is a very high degree of risk involved in trading securities and/or
currencies. The Company, the authors, the publisher, and all affiliates of Company assume no
responsibility or liability for your trading and investment results. Factual statements on the Company's
website, or in its publications, are made as of the date stated and are subject to change without notice.
It should not be assumed that the methods, techniques, or indicators presented in these products will be
profitable or that they will not result in losses. Past results of any individual trader or trading system
published by Company are not indicative of future returns by that trader or system, and are not indicative
of future returns which be realized by you. In addition, the indicators, strategies, columns, articles and all
other features of Company's products (collectively, the "Information") are provided for informational and
educational purposes only and should not be construed as investment advice. Examples presented on
Company's website are for educational purposes only. Such set-ups are not solicitations of any order to
buy or sell. Accordingly, you should not rely solely on the Information in making any investment. Rather,
you should use the Information only as a starting point for doing additional independent research in order
to allow you to form your own opinion regarding investments.
You should always check with your licensed financial advisor and tax advisor to determine the suitability
of any investment.
HYPOTHETICAL OR SIMULATED PERFORMANCE RESULTS HAVE CERTAIN INHERENT
LIMITATIONS. UNLIKE AN ACTUAL PERFORMANCE RECORD, SIMULATED RESULTS DO NOT
REPRESENT ACTUAL TRADING AND MAY NOT BE IMPACTED BY BROKERAGE AND OTHER
SLIPPAGE FEES. ALSO, SINCE THE TRADES HAVE NOT ACTUALLY BEEN EXECUTED, THE
RESULTS MAY HAVE UNDER- OR OVER-COMPENSATED FOR THE IMPACT, IF ANY, OF CERTAIN
MARKET FACTORS, SUCH AS LACK OF LIQUIDITY. SIMULATED TRADING PROGRAMS IN
GENERAL ARE ALSO SUBJECT TO THE FACT THAT THEYARE DESIGNEDWITH THE BENEFIT OF
HINDSIGHT. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO
ACHIEVE PROFITS OR LOSSES SIMILAR TO THOSE SHOWN.
Connors Research
10 Exchange Place
Suite 1800
Jersey City, NJ 07302

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |4

Table of Contents
Section1Introduction.............................................................................5
Section2ShortingMechanics..................................................................8
Section3StrategyRules.........................................................................11
Section4TestResults............................................................................18
Section5SelectingStrategyParameters...............................................24
Section6UsingOptions.........................................................................28
Section7AdditionalThoughts...............................................................31
Appendix:TheConnorsRSIIndicator.....................................................33

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |5

Section1

Introduction

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |6

Formorethanthirtyyears,theassetmanagementindustryhasattemptedtocreatefirmsandfunds
revolvingaroundshortsellingstocks.Few(veryfew)havesucceededinthelongterm.Mostofthese
funds(manynowdefunct)appliedasinglemindedapproachtoshortselling:identifyacompanythatis
goingoutofbusinessandshortit.
Forexample,awellknownshortstrategyistofindthefirmsthatarecommittingaccountingfraudand
shortthem.Inthemid90sIpaid$10,000foraoneyearsubscriptiontoaresearchfirmthatspecialized
inidentifyingaccountingfraud.Thereportthatsubscribersreceivedspelledoutindetailthedepthof
theaccountingfraudafirmwascommitting.Itwascompelling.SocompellingthatofcourseIshorted
manyofthesestocks.Unfortunately,the1995bullmarketbegantotakeholdandinzerocaseswere
anyofthesecompaniesfoundtohavecommittedfraud.Infact,manyofthesestocksexperiencedhigh
doubledigit(andinsomecasestripledigit)gainsoverthenextyear.Fortunately,Iwasoutofallof
thembeforetheymaxedout;Iwaseitherstoppedoutbymyownsenseofpanicorsimplybecausemy
putoptionswenttozero.Overtheyears,thisresearchcompanyidentifiedabout200suspectfirms,of
whichthreeactuallycommittedfraud.Iwasnotahappysubscriber,norwerethemanyshortor
long/shortfundswhoalsosubscribed.Needlesstosay,Ididntrenew.
Thereareotherstrategieswhichhavebeenusedbyshortfundslookingforcompaniestogotozero:
technologiesthatwouldbecomeobsolete,competitorsthatwouldcrushthem,orthefactthattheUS
economywasgoingtocollapseandthereforeeverycompanyalongthewaywouldcollapseaswell.
Noneofthesemethodshaveworkedinthelongterm.Yes,somecompanieshavebeenfoundtohave
committedfraud(Enron,Tyco,etc.),yestheUSeconomygothitin20002002andin2008,butoverall
theperformanceofthemajorityoftheseshortfundshasbeenprettydismal.Thepeoplebehindthese
fundsmeanwell.Mosthaveconvictionintheirbeliefs.Buttheirtrackrecordprovesjusthowdifficultit
istomakemoneyinthelongrunbyshortingstocks.
InthisStrategyGuidebook,weregoingtoteachyouaquantitative,systematicwaytoshortstocksthat
hasproventobesuccessfulforquitesometime.Therearenojuicystorieshere.Thisstrategyissimply
identifyingbehavior,backedbystatistics,whichhasoccurredoverandoveragainsince2001.Infactwe
canshowyouthissamebehaviorgoingallthewaybackto1995.Thebasicbehaviorpatternis:when
stocksbecomeextremelyoverboughtonashorttermbasistheytendtopullbacksharplyforashort
periodoftime.Fewgooutofbusiness.However,mostdopause,profittakingoccurs,scared(long
speculative)moneygetsflushedout,furtherpressureisplacedonthestockasanalystspricetargetsget
hit,andinthemajorityofcases(approximately70%,whichishigh)pricesarelowerwithinafewdays.
Wefirstidentifiedthisbehaviorin2003andstartedteachingavariationofthestrategyinour
TradingMarketsSwingTradingCollegein2005whichwestillteachtoday.Theconceptswetaughtback
thenstillholdtruetodayandwereproudofthefactthatinspiteofmarketschanging,especiallyafter
2008,thebehaviorofextremelyoverboughtstockshasnotchanged.AndthisiswheretheAlphais.
Afewwordsofcautionneedtobemadehere.First,theseareshortsales,whichmeansthatthelosses
arepotentiallyunlimited.Stopslowerthetestresultsbutthatdoesntmeantheyshouldnotbeused;
thatsuptoyou.Thetestresultsincludeallstockswhichhavemettheentrycriteriaforthestrategy,and
This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |7

somehaverisenby50%100%beforeexiting.Onanindividualtradebasisthishurts.Butasyouseeon
anoverallbasis,inspiteoftheseadversemoves,thelongertermtestresultsaresignificantly
positive.Forthoseofyouwhounderstandtherisksinvolvedinoptions,liquidputscanalsobeusedto
containthepotentialforunlimitedlosses,becauseyourrisk(totaldollaramount)isknownaheadof
time.
InordertoachieveAlpha,youhavetogoplacesotherswontgo.Shortingthestocksinthisstrategyisa
placemostpeoplepsychologicallycantgo.Theedgeshavebeenthere,asyouwillseeinthestatistics,
butyouhavetobeabletoovercomethefearofshortingstorystockswhichhavebeenrunupto
unsustainableshorttermlevelsbythecrowd.ThisGuidebook,backedbyoveradecadeofstatistics,
willshowyouhowtodothis.
WehopeyouenjoythisnextinstallmentoftheConnorsResearchStrategyGuidebookSeries.Ifyou
wouldliketoseemoretopicsfromourStrategyResearchSeriespleaseclickhere.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |8

ShortingMechanics
Section2

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |9

Manyindividualinvestorswillnevershortastock.Sometimesthisisbecauseofthefearofassociated
withswimmingagainstthecurrent,asdescribedearlierintheIntroduction.Inothercases,itssimplya
lackofadequateknowledgeabouthowtheprocessworks.Inthissection,wellpresentsomeofthe
basicsthatyoushouldbefamiliarwithbeforeexecutingyourfirstshortsale.
Wereallquiteaccustomedtothemodelwherewepurchasesomethingfirstandsellitlater.Certainly
thisisthecaseformostofthemajorinvestmentswemakeinlife,includinghouses,cars,artwork,
preciousmetals,etc.However,incertainfinancialtransactions,wereallowedtoreversetheprocessby
sellingsomethingbeforewebuyit.Inthecaseofstocks,wecallthisshortselling,orsometimessimply
shorting.
Everystocktransactioninvolvesabuyerwhoiswillingtogiveupcash(eitherherown,orloanedtoher
byabroker)inexchangeforstockheldbytheseller.Sohowdoestheshortsellercompletethis
transactionwithoutactuallyowninganystock?Justasabuyercanuseaccountmargintoeffectively
borrowcashfromabroker,sotoocantheshortsellerborrowsharesfromhisbroker.Ifyourbrokerhas
sharesthattheyarewillingtoloantoyouforshorting,yourtradingplatformwilltypicallydesignate
thosestocksasEasytoBorrow,orsimplyETB.StocksthatarelistedasHardtoBorrow,orHTB,maybe
completelyunavailable,ormayrequireacalltoalivepersonatyourbrokerstradedesk.
Whenweexitashorttrade,wesaythatwearecoveringourposition.Wedothisbybuyingthestock
(i.e.wegiveupcashinexchangeforshares),buttheshareswegetdontstayinouraccount,theygo
backtothebrokertorepaytheloanthatwascreatedwhenweenteredtheshortposition.
Itsworthnotingthatdifferentbrokersmayhavedifferentstocksavailabletoborrow.Veryhighlyliquid
stockswillprobablybeETBfrommostanybroker,butlessliquidstocksmaybeavailablefromsome
brokersandnotothers.Ifyourcurrentbrokerconsistentlydoesntallowyoutoborrowsharesthatyou
wanttoshort,consideropeninganaccountwithanotherbroker.
Becauseyourbrokerisloaningyoushareswhenyouenterashortsale,youmusthaveamarginaccount
withthebrokeragefirm.YoucannotshortstocksfromanIRA,401k,orothercashsecuredaccount.
Also,justlikewhenyoutradeonmargin,thebrokerwillchargeyouinterestonthesharesyouve
borrowed.
Ifyoubuyastockfor$100pershare,yourmaximumriskisthe$100youpaidbecausethestockprice
cannotgolowerthanzero.However,ifyoushortastock,yourriskispotentiallyunlimited,because
theoreticallythereisnoupperboundontheprice.Keepinmindthatifyoureinashortpositionandthe
pricemovesagainstyou,yourbrokerwillrequireyoutohavesufficientcollateral(cashandother
securities)inyouraccounttocoveryourposition.Again,thisisverysimilartohowyourbrokerhandles
longpositionsthatyoupurchasedusingmargin.
Anotherissuetobeawareofisthatshortsellersareresponsibleforpayingdividends.LetssaythatI
borrowastockfrommybroker,andsellitshortat$50/share.Afewdayslater,thecompanypaysa
$1/sharedividendtothenewowner,whichwillcausethestockpricetodropby$1.However,my
brokerstillownssharesaswell,becausetheysimplylentthemtomeforaperiodoftime.Sincethey
This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |10

expecttocollecttheir$1/sharedividendaswell,themoneytopaythatdividendcomesoutofmy
account.Ofcourse,ifIshortedthestockat$50/shareandcoveritafewdayslaterat$49/share,then
Ivemade$1/shareonthetransaction,andthatsthe$1/sharethatpaysthedividend.Inotherwords,
whenacompanypaysadividendandthestockpricefallsbythecorrespondingamount,thereisreally
noneteffectontheshortsellerotherthanthecashcomingoutofhisorheraccountnowversusthe
gainthatwontberealizeduntiltheshortpositioniscovered.
Nowthatwevecoveredsomeofthebasicsaroundshorting,letsmoveontotherulesforthestrategy.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |11

StrategyRules
Section3

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |12

ThefundamentalconceptbehindtheConnorsRSIShortStockStrategyistofindstocksthathavemoved
stronglyupwardandachievednewshorttermhighs.Weshortthesestocksonfurtherintradaystrength,
andthenwaitformeanreversiontopullthepricesbackdownbeforeweexit.
ThisstrategyexecutestradesusingasimplethreestepprocessconsistingofSetup,EntryandExit.The
rulesforeachofthesestepsaredetailedbelow.
ASetupoccurswhenallofthefollowingconditionsaretrue:
1. Thestockspriceclosesabove$5pershare.
2. Theaveragevolumeoverthepast21tradingdays(approximatelyonemonth)isgreater
than500,000shares.
3. ThestockcloseswithaConnorsRSI(3,2,100)valuegreaterthanX,
whereXis75,80,85,90or95.
4. Thestocks100dayHistoricalVolatility,orHV(100),isgreaterthan40.
5. Thestocks10dayAverageDirectionalIndex,orADX(10),valueisgreaterthan40.
6. TodaysHighisthehighesthighinthepastNdays,whereNis7,10or13.
IfthepreviousdaywasaSetup,thenweEnteratradeby:
7. SubmittingalimitordertoshortthestockatapriceY%aboveyesterdaysclose,
whereYis2,4,6,8,or10.
NOTE:SomevariationsuseVariableLimits.Thisconceptwillbeexplainedindetailbelow.
Afterweveenteredthetrade,weExitusingoneofthefollowingmethods,selectedinadvance:
8a. ThestockcloseswithaConnorsRSIvaluelessthan20.
8b. ThestockcloseswithaConnorsRSIvaluelessthan30.
8c. ThestockcloseswithaConnorsRSIvaluelessthan40.
8d. Theclosingpriceofthestockislessthanthe5daymovingaverage,orMA(5).
8e. Theclosingpriceofthestockislowerthanthepreviousdaysclose.Wetypicallyreferto
thisexitastheFirstDownClose.

Letslookateachruleinalittlemoredepth,andexplainwhyitsincludedinthestrategy.
Rules1&2assurethatweretradingliquidstockswhicharelikelytohavesharesavailabletoborrow.
Rule3usesConnorsRSItoidentifyapricesurge.AcompletedescriptionofConnorsRSIcanbefoundin
theAppendix.
Rule4helpsselectstocksthathavebeenexperiencinghealthylevelsofpricevolatilityoverthepast
severalmonths.Shorttermstrategiesliketheonepresentedheredependonstrongpricemovesboth
tosetupthetradeandtogenerateprofits.Stockswithlowervolatilityarelesslikelytogeneratesuch
moves.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |13

Rule5usestheADXindicatortoidentifystocksthathavebeentrendingstrongly.AlthoughADXisnon
directional,theotherrulesassureusthatthepricehasbeenmovingupward,andthusADXmeasures
thestrengthofthatupwardtrend.
Rule6identifiesanNdayhigh.Stocksseldommoveinonedirectionforanextendedperiodoftime,and
whenthepricereachesashorttermhigh,itsoftenasignthatthestockisgettingreadytopullback
beforeresumingitsupwardtrend.
Rule7allowsustoenterthetradeatanoptimalprice.TheSetuprulesidentifyanoverboughtstock,
andtheentryrulewaitsforittobecomeevenmoreoverboughtonanintradaybasis.Becausethe
intradaypricesurgeisoccurringforasecondconsecutiveday,itoftenmotivatestraderswithlong
positionstotakesomeprofitsbyclosingtheirpositions.Theirsellingwillgeneratedownwardpressure
ontheprice,whichinturnwillgenerateprofitsforourshortposition.
Overtheyears,ourresearchhasconsistentlyshownthatstocksthataretradingabovetheir200day
movingaverage,orMA(200),haveagreatertendencytomoveupwardinprice,whilestocksbelowtheir
MA(200)haveagreatertendencytomovedownward.UsingVariableLimitEntriesallowsustoleverage
thisinformationbyspecifyingalargerlimitorderwhenthestockisabovetheMA(200).
StrategyvariationsthatuseVariableLimitEntriesusealimitpercentagethatis1.5timesthestandard
limitwhenthestockisabovetheMA(200).Forexample,ifwereusingavariationwithastatedlimit
percentage(Y)of4%,thenastockthatclosedabovetheMA(200)ontheSetupdaywillactuallyusea
6%limitordertoenterthetrade,whileastocktradingbelowtheMA(200)willusethe4%limit.Strategy
variationsthatdonotutilizeVariableLimitEntriesusethesamelimitpercentageregardlessofwhether
thepriceontheSetupdayclosesaboveorbelowtheMA(200).Wewillseeexamplesofbothkindsof
tradeslaterinthissection.
Rule8providesawelldefinedexitmethod.Fewstrategieshavequantified,structured,anddisciplined
exitrules.Rule8givesyoutheexactparameterstoexitthetrade,backedbyovertwelveyearsof
historicaltestresults.Aswithallotherstrategyparameters,weselectinadvancethetypeofexitthat
wewilluse,andapplythatruleconsistentlyinourtrading.
InourtestingweclosedalltradesatthecloseoftradingonthedaythattheExitsignaloccurred.Ifthisis
notanoptionforyou,ourresearchhasgenerallyshownthatsimilarresultsareachievedifyouexityour
positionsthenextmorning.
Nowletsseehowatypicaltradelooksonachart.Fortheexamplebelow,welluseastrategyvariation
thatrequirestheConnorsRSIvaluetobeabove90andthatrequiresa10dayhighontheSetupday.The
limitorderwillbeplaced6%abovetheSetupdaysclosingprice(novariablelimits).Wewillexitwhen
ConnorsRSIclosesbelow30.Intermsofourstrategyrulesabove,thatmeansX=90,N=10,Y=6,and
theexitmethodisdefinedbyRule8b.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |14

ChartcreatedinAmibroker.ReprintedcourtesyofAmiBroker.com.

Figure1:ZillowTrade

ThechartaboveisforZillowInc.,whosesymbolisZ.Inthechart,thetoppaneshowsthepricebarsin
blackandthe200daymovingaverageorMA(200)indarkgreen.Theverticalbluegraylinemarksthe
currentlyselecteddaywhichisalsotheSetupday,February14,2013.Thereddownarrowmarksthe
entryday,andthegreenuparrowindicatestheexitday.Themiddlepaneshowsthevaluesfor
ConnorsRSIinbrightblue,100dayhistoricalvolatilityorHV(100)inpurpleandADX(10)inbrightgreen.
Thebottompaneshowsthedailyvolumeasabluegrayhistogram,andthe21dayaveragevolumeasa
turquoiseline.Nowwellconfirmthateachofourentryandexitconditionswascorrectlymet.
Rule1requiresthestockpricetocloseabove$5ontheSetupday.Theclosingpriceof$42.30meets
thisrequirement.
Rule2issatisfiedbecausethe21dayaveragevolumeisover725,000shares,whichiswellaboveour
minimumrequirementof500,000shares.
Basedonourstrategyparameters,Rule3requirestheConnorsRSI(3,2,100)valuetobeabove90onthe
Setupday,whichitis:thevalueshownonthechartontheSetupdayis90.33.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |15

Rule4issatisfiedbecausetheHV(100)valueof58.56ontheSetupdayisabovethethresholdof40.
Rule5isalsofulfilledbecausetheADX(10)valueof54.05isabovethethresholdof40.
Rule6requiresanNDayhigh,andforthisstrategyvariationN=10.TheSetupdaycountsasDay1of
the10daylookbackperiod.Countingbackanadditional9dayswilltakeyoutotheleftedgeofthe
chart.YoucanseethatthehighontheSetupdayishigherthananyoftheotherhighsduringthe
lookbackperiod,andthustheRule6criteriahavebeenmet.
SinceallthreeSetupruleshavebeensatisfied,weenteralimitorderforthenexttradingday,whichis
February15th.Ourselectedstrategyvariationtellsustousealimitof6%abovetheSetupdaysclosing
price,sowewouldusealimitpriceof:

LimitPrice
=Closex(1+Limit%)

=$42.30x1.06=$44.84

OnFebruary15ththepriceofZclimbsto$45.38,soourlimitordergetsfilledandweshortthestockat
thelimitpriceof$44.84.
Onthenexttradingday,February19th,thepriceofZopenslowerbutcloseswithapriceverynearthe
entrydaysclosingprice.ThisbringstheConnorsRSIvaluedowntoabout50.Sincethisisnotbelowour
exitthresholdof30,westayinthetrade.
OnFebruary20th,thepricefallsfurther.ThistimetheclosingvalueofConnorsRSIis25.22,whichsignals
ourexit.Wecoverourpositionatorneartheclosingpriceof$42.65,whichgivesusaniceprofitonthe
tradeofjustunder5%beforecommissionsandfees:

Profit

=Gain(orLoss)/CostBasis

=($44.84$42.65)/$44.84

=$2.19/$44.84=4.9%

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |16

Letslookatanotherexampleusingthesamestrategyparameters,exceptthistimewithvariablelimits
enabled.ThechartbelowisforOsirisTherapeutics(OSIR),andusesthesameconventionsasthe
previouschart.

ChartcreatedinAmibroker.ReprintedcourtesyofAmiBroker.com.

Figure2:ORISTradeusingVariableLimits

TheSetupdayforthistradewasJune21,2012.AsperRule1,OSIRistradingabove$5andasperRule2
the21dayaveragevolumeisabove500,000shares.Thusourliquidityrequirementshavebeenmet.
Rule3issatisfiedbyaConnorsRSIvalueof93.15(greaterthan90)ontheSetupday,Rule4istakencare
ofbyanHV(100)valueof76.22(greaterthan40),andRule5iscoveredbyanADX(10)valueof51.90
(alsogreaterthan40).

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |17

AsperRule6,wecanveryclearlyseethatOSIRhasmadea10dayHigh.Thissteadyupwardmarchover
atwoweekperiodisclassicbehaviorfortradecandidateswhenusingtheConnorsRSIShortStock
Strategy.
WithallofourSetupconditionsmet,wearereadytoplacealimitorderforthenextday.Sincetheprice
ofOSIRisabovetheMA(200)ontheSetupday,wewillusealimitorderof9%insteadof6%,asper
Rule7.Thatmakesourlimitprice$11.59x1.09=$12.63.
OnJune22nd,thepriceofOSIRhitsanintradayhighof$12.75,whichisaboveourlimitprice,soour
ordergetsfilledandweenterthetrade.
Forthenexttwodays,thepriceofOSIRcontinuestorise,andtheConnorsRSIvaluestaysnearthetop
endoftherange.Then,onJune27th,thepricedropsbyover17%inasingleday,pullingConnorsRSI
downto15.2andsignalingourexitasperRule8.Wecoverourshortpositionatorneartheclosing
priceof$11.42.Thistradewouldhavegeneratedaprofitofapproximately9.6%beforecommissions
andfees.
Nowthatyouhaveagoodunderstandingofthetrademechanics,welllookatthehistoricaltestresults
fordifferentvariationsofthestrategy.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |18

Section4

TestResults

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |19

Wecanneverknowforsurehowatradingstrategywillperforminthefuture.However,forafully
quantifiedstrategysuchastheonedescribedinthisGuidebook,wecanatleastevaluatehowthe
strategyhasperformedinthepast.Thisprocessisknownasbacktesting.
Toexecuteabacktest,wefirstselectagroupofsecurities(sometimescalledawatchlist)thatwewant
totestthestrategyon.Inourcase,thewatchlistconsistsofpastandpresentmembersoftheS&P500.
Nextwechooseatimeframeoverwhichtotest.Thelongerthetimeframe,themoresignificantand
informativethebacktestingresultswillbe.ThebacktestsforthisGuidebookstartinJanuary2001and
gothroughtheendofApril2013,thelatestdateforwhichwehavedataasofthiswriting.
Finally,weapplyourentryandexitrulestoeachstockinthewatchlistfortheentiretestperiod,
recordingdataforeachtradethatwouldhavebeenentered,andaggregatingalltradedataacrossa
specificstrategyvariation.
OneofthekeystatisticsthatwecangleanfromthebacktestedresultsistheAverage%Profit/Loss,also
knownastheAverageGainperTrade.Sometradersrefertothisastheedge.TheAverage%P/Listhe
sumofallthegains(expressedasapercentage)andallthelosses(alsoasapercentage)dividedbythe
totalnumberoftrades.Considerthefollowingtentrades:
TradeNo.
1
2
3
4
5
6
7
8
9
10

%GainorLoss
1.7%
2.1%
4.0%
0.6%
1.2%
3.8%
1.9%
0.4%
3.7%
2.6%

TheAverage%P/Lwouldbecalculatedas:
Average%P/L=(1.7%+2.1%4.0%+0.6%1.2%+3.8%+1.9%0.4%+3.7%+2.6%)/10
Average%P/L=1.08%
Average%P/Listheaveragegainbasedoninvestedcapital,i.e.theamountofmoneythatweactually
spenttoentereachtrade.
Forshorttermtradeslastingthreetotentradingdays,mosttraderslookforanAverage%P/Lof0.5%
to2.5%acrossalltrades.Allotherthingsbeingequal,thelargertheAverage%P/L,themoreyour
accountwillgrowovertime.Ofcourse,allotherthingsareneverequal!Inparticular,itsimportantto
considertheNumberofTradesmetricincombinationwithAverage%P/L.Ifyouuseapproximatelythe
This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |20

sameamountofcapitalforeachtradethatyouenter,youllmakealotmoremoneyontentradeswith
anaverageprofitof4%pertradethanyouwillononetradethatmakes10%.
AnotherimportantmetricistheWinningPercentageorWinRate.Thisissimplythenumberof
profitabletradesdividedbythetotalnumberoftrades.Inthetableabove,7ofthe10tradeswere
profitable,i.e.hadpositivereturns.Forthisexample,theWinningPercentageis7/10=70%.
WhydowecareaboutWinRate,aslongaswehaveasufficientlyhighAverage%P/L?Becausehigher
WinRatesgenerallyleadtolessvolatileportfoliogrowth.Losingtradeshaveawayofclumpingup,
andwhentheydothat,thevalueofyourportfoliodecreases.Thisisknownasdrawdown.Those
decreases,inturn,canmakeyoulosesleeporevenconsiderabandoningyourtradingaltogether.If
therearefewerlosers,i.e.ahigherWinningPercentage,thenlossesarelesslikelytoclump,andyour
portfoliovalueismorelikelytogrowsmoothlyupwardratherthanexperiencingviolentupanddown
swings.
***

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |21

LetsturnourattentiontothetestresultsforthedifferentvariationsoftheShortSellingStockswith
ConnorsRSIStrategy.First,wellsortthetestresultstoshowthe20variationsthatproducedthehighest
Average%P/L.Allvariationsthatgeneratedfewerthan200tradesignalsduringthe12+yeartesting
periodhavebeenfilteredouttoavoidskewingtheresults.
Top20VariationsBasedonAverageGain

#
Trades
203
339
244
319
357
205
273
205
206
493
382
246
576
419
343
382
206
343
435
364

Avg
%P/L
9.40%
7.42%
7.08%
6.97%
6.88%
6.85%
6.55%
6.41%
6.28%
6.21%
6.19%
5.89%
5.88%
5.82%
5.79%
5.76%
5.73%
5.70%
5.61%
5.55%

Avg
Days
Held
10.24
10.29
11.49
10.70
11.55
3.74
11.37
3.79
2.19
11.69
11.01
3.88
11.73
10.99
4.04
11.35
1.57
3.84
11.15
4.13

Win Entry
Rate CRSI
78.82% 90
76.11% 90
75.00% 90
75.55% 90
73.11% 85
77.56% 90
75.82% 90
76.10% 90
76.70% 90
70.99% 80
74.08% 90
76.02% 90
69.79% 75
74.46% 90
76.97% 90
73.82% 90
71.84% 90
76.09% 90
74.71% 90
75.00% 85

#Daysfor
Highest
High
13
13
10
13
13
13
7
13
13
13
10
10
13
7
13
10
13
13
7
13

Var.
Entry
Limit% Limits
10
Y
10
N
10
Y
8
Y
10
Y
10
Y
10
Y
10
Y
10
Y
10
Y
10
N
10
Y
10
Y
10
N
10
N
8
Y
10
Y
10
N
8
Y
10
Y

ExitMethod
CRSI<20
CRSI<20
CRSI<20
CRSI<20
CRSI<20
CRSI<30
CRSI<20
Close<MA(5)
CRSI<40
CRSI<20
CRSI<20
CRSI<30
CRSI<20
CRSI<20
CRSI<30
CRSI<20
DownClose
Close<MA(5)
CRSI<20
CRSI<30

Belowisanexplanationofeachcolumn.
#TradesisthenumberoftimesthisvariationtriggeredfromJanuary1,2001April30,2013.
Avg%P/Listheaveragepercentageprofitorlossforalltrades,includingthelosingtrades,basedon
investedcapital.Thetop20variationshaveallshownpositivegainsrangingfrom5.55%to9.4%over
the12+yeartestingperiod.
AvgDaysHeldistheaveragetradedurationexpressedasanumberofdays.Therangeforthevariations
aboveisrelativelylarge,fromjustover2daystoovertwoweeks(11+tradingdays).Wewilldiscussthis
furtherinalatersection.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |22

Win%isthepercentageofsimulatedtradeswhichclosedoutataprofit.Mostofthetop20variations
havewinratesinthemid70s.Thisisahighpercentageofprofitabletradesinaworldwheremany
tradersareaimingfor5060%.
EntryCRSIcorrespondstoRule3ofthestrategy,whichstatesthattheConnorsRSIvaluemustbeabove
theentrythreshold.RecallthatwetestedwithConnorsRSIthresholdsof75,80,85,90and95.Asyou
mightexpect,thehigherConnorsRSIvaluesdominatethelist,althoughthe95thresholdissuspiciously
absent.Again,wewillexplorethisfurtherinalatersection.
#DaysforHighestHighcorrespondstoRule6ofthestrategy.Againweseeapreponderanceofhigher
valuesinthelist,demonstratingthatwecanoftenachievelargergainsbywaitingformorerestrictive
entrycriteriatobemet.
Limit%isrelatedtoRule7ofthestrategy,anddeterminesthelimitpricethatwillbeusedtoenterthe
trade.Wetestedlimitsof2,4,6,8and10%abovetheSetupdaysclose.
Var.EntryLimitsisYes(Y)whenthetestusedanentrylimitof1.5timesnormalforstocksthatwere
abovetheMA(200).Whenthesamelimitwasusedregardlessofwhetherthepricewasaboveorbelow
theMA(200),thiscolumncontainsaNo(N).WeseemanymoreYsthanNs,indicatingthatusing
variableentrylimitswasbeneficial.
ExitMethodistherulethatwasusedtoexittradesinthisstrategyvariation,asdescribedinRule8.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |23

Next,letslookatthestrategyvariationsthathavehistoricallyhadthehighestfrequencyofprofitable
tradesorWinRate.
Top20VariationsBasedonHighestWinRate

#
Trades
203
205
343
206
266
321
282
310
339
205
343
246
279
386
344
323
273
333
275
328

Avg
%P/L
9.40%
6.85%
5.79%
6.28%
4.42%
5.47%
4.37%
4.00%
7.42%
6.41%
5.70%
5.89%
4.32%
5.26%
5.37%
4.77%
6.55%
3.91%
5.37%
3.85%

Avg
Days
Held
10.24
3.74
4.04
2.19
3.70
3.99
3.74
3.68
10.29
3.79
3.84
3.88
3.76
4.09
2.24
2.25
11.37
3.73
3.96
3.74

Win Entry
Rate CRSI
78.82% 90
77.56% 90
76.97% 90
76.70% 90
76.69% 95
76.64% 90
76.24% 95
76.13% 95
76.11% 90
76.10% 90
76.09% 90
76.02% 90
75.99% 95
75.91% 90
75.87% 90
75.85% 90
75.82% 90
75.68% 95
75.64% 90
75.61% 95

#Daysfor
Highest
High
13
13
13
13
13
13
7
13
13
13
13
10
10
10
13
13
7
7
7
10

Var.
Entry
Limit% Limits
10
Y
10
Y
10
N
10
Y
6
N
8
Y
6
N
4
Y
10
N
10
Y
10
N
10
Y
6
N
10
N
10
N
8
Y
10
Y
4
Y
10
Y
4
Y

ExitMethod
CRSI<20
CRSI<30
CRSI<30
CRSI<40
Close<MA(5)
CRSI<30
Close<MA(5)
Close<MA(5)
CRSI<20
Close<MA(5)
Close<MA(5)
CRSI<30
Close<MA(5)
CRSI<30
CRSI<40
CRSI<40
CRSI<20
Close<MA(5)
CRSI<30
Close<MA(5)

All20ofthetopvariationshavehistoricallyproducedaprofitonover75%oftheidentifiedtrades!
Noticethatthereisagooddealofoverlapbetweenthislistandtheonepresentedintheprevious
sectiononAverage%P/L,tellingusthatwehavemultiplestrategyvariationsthathavehistoricallywon
consistentlywhileproducingexcellentedges.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |24

Section5

SelectingStrategy
Parameters

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |25

Inpreviouschapterswevedescribedthedifferentvalueswetestedforstrategyparameterssuchasthe
ConnorsRSIentrythreshold(X),NDayHigh,entrylimit%(Y)andexitmethod.Inthissectionwell
discusssomeadditionalthingstoconsiderasyoudecidewhichvariation(s)touseinyourtrading.
Letstalkconceptuallyaboutentriesandexitsforamoment.Bothentryandexitrulescanbethoughtof
intermsofhowstricttheyare,i.e.howeasyordifficulttheyaretoachieve.Youmightalsosaythat
strictnessisameasureofhowfrequentlyorinfrequentlytheruleconditionsoccur.Foroscillatorssuch
asConnorsRSI,valuesthatareclosertotheextremes(0and100)aremorestrict(lesslikelytooccur)
thanvaluesthatareinthemiddleoftherange.
Stricterentryruleswillbesatisfiedlessfrequentlythanmoreleniententryrules,andthusastrategy
thatreliesonthestricterruleswillgenerallygeneratefewertradesthanastrategywhoseentryrulesare
moreeasilysatisfied.Witharobuststrategy,therewardforfewertradesisusuallyahighergainper
trade,onaverage.Ifyoushortaslightlyoverboughtstock,itsmostlikelytohaveamoderatepullback.
Butifyouwaitforthestocktobecomeextremelyoverbought,thechancesaremuchhigherthatitwill
experienceasignificantpricedropandcreateabiggerprofit.
Incontrasttoentryrules,thestrictnessofexitruleshaslittleeffectonthenumberoftradesgenerated
bythestrategy.However,justliketheentryrules,stricterexitrulestypicallyresultinhigheraverage
profits.Why?Becausestricterexitrulestendtokeepyouinyourtradesforalongertime,givingthe
stockmoretimetoexperiencethemeanreversionbehaviorthatwereattemptingtoexploitwitha
strategyliketheShortSellingStockswithConnorsRSIStrategy.Thus,forentriesthetradeoffisbetween
moretradesandhighergainspertrade,whileforexitsthetradeoffisbetweenshortertradedurations
andhighergainspertrade.
***

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |26

NowletsturnourattentionbacktothestrategydescribedinthisGuidebook.Inthetablebelow,we
comparefivevariationsofthestrategythatallusethesamenumberofdays(10)forthehighesthigh,
thesamestaticlimitentry(6%)andthesameexitmethod(ConnorsRSI<30).OnlytheConnorsRSIentry
thresholddiffersbetweenthevariationsshownbelow.
TheEffectofConnorsRSIEntryThreshold

#
Trades
3209
2635
1799
898
279

Avg
%P/L
2.41%
2.55%
2.76%
3.47%
4.47%

Avg
Days
Held
4.69
4.66
4.53
4.27
4.00

Win Entry
Rate CRSI
69.43% 75
69.56% 80
70.71% 85
71.71% 90
74.19% 95

#Daysfor
Highest
High
10
10
10
10
10

Var.
Entry
Limit% Limits
6
N
6
N
6
N
6
N
6
N

ExitMethod
CRSI<30
CRSI<30
CRSI<30
CRSI<30
CRSI<30

Noticethatthemostleniententryinthetable,thefirstlinewithaConnorsRSIEntryThresholdof75,
generatedthemosttradesignalsandthelowestgainpertrade.Astheentryrulebecomesstricter,i.e.
theConnorsRSIEntryThresholdrises,weseefewerandfewertradesignalsbuthigherandhigher
averagegainspertrade.Thevariationwithanentrythresholdof95hasnearlydoubletheAverage%P/L
asthefirstvariation,butalsohaslessthan1/10ththenumberoftrades.
Itshouldcomeasnosurprisethatthepatternemergesagainwhenweholdallparametersconstant
excepttheLimit%usedtodeterminethelimitentryprice.IfwekeeptheSetupconditionsconstant,
thentherewillobviouslybemorestocksthatexperienceapricesurgeof2%orgreaterthenextday
thantherewillbethosethatrisebyatleast10%.
VariationswithDifferentLimit%Entries

#
Trades
5422
3078
1799
1119
745

Avg
%P/L
1.41%
2.13%
2.76%
3.23%
4.09%

Avg
Days
Held
4.77
4.64
4.53
4.52
4.44

Win Entry
Rate CRSI
67.54% 85
69.36% 85
70.71% 85
71.58% 85
73.42% 85

#Daysfor
Highest
High
10
10
10
10
10

Var.
Entry
Limit% Limits
2
N
4
N
6
N
8
N
10
N

ExitMethod
CRSI<30
CRSI<30
CRSI<30
CRSI<30
CRSI<30

Wehaveconfirmedthatstricterentryrulesresultinfewertradesbuthigheraveragegains.Nowlets
lookattheexits.Hereweholdthesetupandentrycriteriaconstant,butvarytheexitmethods:

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |27

VariationswithDifferentExitMethods

#
Trades
1834
1825
1803
1799
1743

Avg
%P/L
2.36%
2.62%
2.64%
2.76%
2.95%

Avg
Days
Held
1.75
2.56
4.07
4.53
11.76

Win Entry
Rate CRSI
68.70% 85
70.08% 85
70.55% 85
70.71% 85
70.22% 85

#Daysfor
Highest
High
10
10
10
10
10

Var.
Entry
Limit% Limits
6
N
6
N
6
N
6
N
6
N

ExitMethod
DownClose
CRSI<40
Close<MA(5)
CRSI<30
CRSI<20

Aswepredictedearlier,allfivevariationsgeneratedverysimilarnumbersoftradesignals.Therangeis
from1743tradesto1834trades,whichrepresentsonlyabouta+/3%variationfromtheaveragevalue
of1801.However,thevariationthatusesthemostlenientexitmethod(coveringthepositiononthe
firstdaythatthestockpriceclosesdown)generatesanaveragegainthatisonly80%ofthestrictestexit
method.
Evenmorenotableisthedifferenceinthetradeduration,whichrangesfrom1.75daysto11.75days!
Itsalsointerestingtocomparethesecondandthirdlinesinthetable.Usinganexitmethodof
ConnorsRSI<40generated1825tradesignalswithanaveragegainpertradeof2.62%andanaverage
durationof2.56days.UsinganexitmethodofClose>MA(5)generatednearlythesamenumberof
trades(1803)andaveragegain(2.64%),butwithanaveragetradedurationofover4days.Ifyouwere
selectingbetweenthesetwovariationsonly,itwouldobviouslybebettertousethefirstvariationand
takeyourprofitsmorequickly.Inthisinstance,holdingthetradeforalongertimeproducedverylittle
incrementalbenefit.
Armedwiththisinformation,youwillnowbeabletoselectstrategyparametersthataremostlikelyto
producethenumberoftradesignals,averagegains,andtradedurationthatbestcomplementyour
overalltradingplan.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |28

UsingOptions
Section6

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |29

Optionstradinghasbeenamajorgrowthindustryoverthepastseveralyearsinthemarkets.Thisis
becausespreadshavetightened,liquidityhasincreased,andtheabilitytoeasilytradecomplexoptions
hasneverbeensimpler.
Wellnowfocusonapplyingoptionstradingtotheshorttermmarketmoveswehavejustlearned.Like
everythingelseinthisGuidebook,therearedefinitiverulesastohowtoexecuteanoptionstradewhen
astrategysignaltriggers.
Hereiswhatweknowbaseduponthedata:
1.Themajorityofthemovesfromentrytoexithavebeenheldaveryshortperiodoftime(212
tradingdays).
2.Theaveragegainspertradehavebeenlargewellbeyondthenormaldistributionofprices
overthatshortperiodoftime.
3.Ahighpercentageofthemoveshavebeendirectionallycorrect.
Whenwelookatthistypeofbehavior,itcanleadtomanystrategiesbutonestrategystandsout(and
thishasbeenconfirmedbyprofessionaltraders).Thestrategyistobuyfrontmonth,inthemoneylong
puts.
Whyfrontmonth,inthemoneylongputs?Becausetheywillmovemostcloselytothestockitself.And
thecloseranoptionmoveswiththestock,thegreaterthegainwillbeonapercentagebasiswhenthe
moveiscorrect.
Herearetherules.
1.Asignaltriggers.
2.Buythefrontmonthinthemoneyput.Ifyouweretonormallyshort500sharesofthestock,
buy5puts(every100sharesshouldequaloneputoptioncontract).
3.Exittheoptionswhenthesignaltriggersanexitonthestock.
Letsgofurther:
1.Whatdoesinthemoneyexactlymeanhere?
Inthiscaseitsdefinedasonetotwostrikepricesinthemoney.Ifthestockpriceisat48andthe
intervalbetweenoptioncontractsis$5,thenbuythe50or55puts.
2.Whatdoesfrontmonthmean?
Becausetheholdingperiodissoshort,youwanttotradetheoptionswhosemonthlyexpirationisthe
closest.Iftheclosestmonthiseighttradingdaysorlessfromthefrontmonthsoptionexpirationdate
(meaningthesecondWednesdaybeforeorcloser)usethefollowingmonthastheonetotrade.
This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |30

3.WhathappensifIminthepositionanditexpires,yetthesignalforthestockisstillvalid?
Inthiscase,rolltothenextmonth.Youretradingthestocksignalssoyouwanttohaveexposuretothat
signal.
4.Whataboutliquidityandspreads?
Theressomediscretionhere.Thereisnohardandfastruleastowhatexactlyliquiditymeansin
options.Manytraderslookforminimumvolumeand/oropeninteresttodetermineliquidity.
Assumingthereisactivevolumeintheoptions,lookatthespreads.Iftheoptionistrading3.00bid/
3.30offer,thespreadis10%.Canyoureallyovercomea10%spread?Notlikely.Nowcomparethisto
anoptionthatstradingat3.25bid/3.30offer.Thisisfarmoreacceptableandtradable.
5.Whataretheadvantagesofbuyingputoptionsinsteadofshortingthestock?
Assumingthespreadsandliquidityarethere,theadvantagesarelarge:
1. GreaterpotentialROIoncapitalinvested.
2. Lessmoneytiedup.
3. Lesspointsatrisk.Thismeansifyoushortastockat50,yourlossesaretheoretically
unlimited.Theoptionscanonlyloseuptothepremiumyoupaid.So,ifyouboughtthe55
puts,theriskisonlythepremium.
4. Theresgreaterflexibility.Forexample,letssaythestocktriggeredashortsignalat50and
youpaid$5.50forthe55puts.Ifthestockimmediatelymoveslower(letssayto46);you
havechoiceshere.Youcanexit,oryoucanrollintothe50putgettingmostofyourmoney
outandnowturningthisintoanearlyfreetradeifyoubelievethatpriceswillcontinueto
fall.
Therearenumerousexampleslikethisandyoucanfindthesetypesofstrategyopportunitiesinmost
optionsbooks.Buttradinganythingexoticordifferentthansimplybuyingtheputsisagainsttheadvice
ofthemanyprofessionalsweposedthisquestionto.
Inconclusion,optionsprovidetraderswithagoodalternativetoshortingthestockoutright.The
structuredmethodologyforourstrategiesis:frontmonth,inthemoney,withequivalentsizing(1
optionper100shares),andexitingwhenthesignalexits.
Theaboveoptionsstrategy,inmanyexpertsopinion,isthebestandmostefficientstrategybasedupon
thehistoricaldatafromthesesignals.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |31

AdditionalThoughts
Section7

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |32

1.AsyouhaveseenthroughoutthisGuidebook,theShortSellingStockswithConnorsRSIStrategyhas
hadlargequantifiededgeswhenappliedinasystematicmanner.
2.Thereareliterallyhundredsofpotentialvariationsforyoutouse.Byadjustingtheinputvariables
describedintherules,youcancustomizehowthestrategywillperformforyou.Wantmoretrades?
LookatvariationswithalowerConnorsRSIentryvalueorfewerlookbackdaysforthehighesthigh.
Biggeraveragereturns?Checkoutthevariationsthathavethestrictestentrycriteria(highentryvalue
forConnorsRSIandhighLimit%)andlongestdurations(ConnorsRSI20exitmethod).Wanttogetinand
outoftradesmorequicklytoreduceovernightriskandfreeupyourcapitalforothertrades?Trythe
variationsthatutilizetheFirstDownCloseexitmethod.
3.Whataboutstops(andweincludetheanswertothisinallourStrategyGuidebooks)?
WehavepublishedresearchonstopsinotherpublicationsincludinginourbookShortTermTrading
StrategiesThatWork.
Whatwehavefoundisthatstopstendtolessenperformanceandinmanycasestheycompletely
removeedges.Yes,itfeelsgoodwhenastockkeepsmovinglowerandlowerandastopgotyouout.On
theotherside,theresearchwhichisbackedbyuptotwodecadesoftestresultsonmanyshortterm
tradingstrategiessuggeststhatstopsgethitoftenandaccumulatemany,manylosses.Fewtrading
strategiescanovercometheseaggregatedlosses.
Formanytradersstopsareamust.Psychologicallyitallowsthemtotaketrades,especiallydifficult
trades.Whetheryouusethemornotisapersonalchoice.Onthewholethough,theedgesyouseein
thisstrategyandmanyothershorttermstrategiesarelowerwhenstopsareappliedtothem.Againthis
isapersonalchoiceonlyyoucanmakeforyourself.Weknowsuccessfultradersinbothcamps.
4.Slippageandcommissionwerenotusedinthetesting.Factorthemintoyourtrading(theentriesare
atlimitpricessoslippageisnotanissue)andmakesureyouretradingatthelowestpossiblecosts.
Mostfirmsarenowallowingtraderstotradeforunder1centashare,soshopyourbusiness,especially
ifyouareanactivetrader.Theonlinebrokeragefirmswantyourbusiness.
WehopeyouenjoyedthisadditiontotheConnorsResearchTradingStrategySeries.Ifyouhaveany
questionsaboutthisstrategypleasefeelfreetoemailusatinfo@connorsresearch.com

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |33

Appendix:

TheConnorsRSI
Indicator

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |34

LarryConnorsandConnorsResearchhavebeendeveloping,testing,andpublishingquantifiedtrading
strategiessincethemid1990s.Duringthattime,wehavehadtheopportunitytoevaluateagreat
numberofdifferenttechnicalindicatorsandtoassesstheireffectivenessinpredictingfutureprice
action.Nowwevetakenthenextstepandcreatedanindicatorofourown:ConnorsRSI.Inthischapter
wewilldescribetheindicatorandprovidedetailsonitscalculation.
ConnorsRSIisacompositeindicatorconsistingofthreecomponents.Twoofthethreecomponents
utilizetheRelativeStrengthIndex(RSI)calculationsdevelopedbyWellesWilderinthe1970s,andthe
thirdcomponentranksthemostrecentpricechangeonascaleof0to100.Takentogether,thesethree
factorsformamomentumoscillator,i.e.anindicatorthatfluctuatesbetween0and100toindicatethe
leveltowhichasecurityisoverbought(highvalues)oroversold(lowvalues).
BeforewediscusshowtocalculateConnorsRSI,letsreviewWildersRSI.RSIisaveryusefulandpopular
momentumoscillatorthatcomparesthemagnitudeofastock'sgainstothemagnitudeofitslossesover
somelookbackperiod.Wilderhimselfbelievedthat14periodswastheideallookback.Weoftenuse
theshorthandnotationRSI(14)forthe14periodRSI.TheformulabelowcomputesRSI(14)foraseriesof
pricechanges:

IfwewantedtocomputeRSIforadifferentnumberofperiods(N),thenwewouldreplace14inthe
formulaabovewithN,andreplace13withN1.Regardlessofthenumberofperiodsusedinthe
calculation,theresultwillalwaysbeanumberbetween0and100.TraderswhouseRSI(14)typically
lookforvaluesgreaterthan70toidentifyoverboughtconditions,andvalueslessthan30toindicate
oversoldconditions.
OurpreviousresearchhasshownthatusingshorterlookbackperiodsmakesRSImoreeffectivein
predictingshorttermpricemovements.WehavepublishedmanystrategiesthatutilizeRSI(2),aswellas
severalthatuseRSI(3)andRSI(4).ChangingthenumberofperiodsalsohasaneffectontheRSIlevels
This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |35

thatbestidentifyoverboughtandoversoldconditions.Forexample,anRSI(2)valueoflessthan10is
usuallyareliableindicatorofanoversoldcondition,whileanRSI(2)valueover90isagoodbenchmark
foranoverboughtcondition.
NowletsturnourattentionbacktoConnorsRSI.Asmentionedpreviously,ConnorsRSIcombinesthree
components,andasyoumightguess,theyareallelementsthatourresearchhasrepeatedlyshownto
havesignificantpredictiveability:
PriceMomentum:Aswejustdiscussed,RSIisanexcellentwaytomeasurepricemomentum,
i.e.overboughtandoversoldconditions.Bydefault,ConnorsRSIappliesa3periodRSI
calculationtothedailyclosingpricesofasecurity.WewillrefertothisvalueasRSI(Close,3).
DurationofUp/DownTrend:Whentheclosingpriceofasecurityislowertodaythanitwas
yesterday,wesaythatithascloseddown.Ifyesterdaysclosingpricewaslowerthanthe
previousdaysclose,thenwehaveastreakoftwodownclosedays.Ourresearchhasshown
thatthelongerthedurationofadownstreak,themorethestockpriceislikelytobouncewhen
itrevertstothemean.Likewise,longerdurationupstreaksresultinlargermovesdownwhen
thestockmeanreverts.Ineffect,thestreakdurationisanothertypeofoverbought/oversold
indicator.
Theproblemis,thenumberofdaysinastreakistheoreticallyunbounded,thoughwecould
probablyplacesomepracticallimitsonitbasedonpastexperience.Forexample,wemight
observethattherehavebeenveryfewinstancesofeitheranupstreakoradownstreaklasting
formorethan20days,butthatstilldoesntgetustoatypicaloscillatortypevaluethatvaries
between0and100.
Thesolutionistwofold.First,whenwecountthenumberofdaysinastreak,wewilluse
positivenumbersforanupstreak,andnegativenumbersforadownstreak.Aquickexample
willhelptoillustratethis:
Day
1
2
3
4
5
6
7
8

ClosingPrice
$20.00
$20.50
$20.75
$19.75
$19.50
$19.35
$19.35
$19.40

StreakDuration

1
2
1
2
3
0
1

TheclosingpriceonDay2ishigherthanonDay1,sowehaveaonedayupstreak.OnDay3,
thepricecloseshigheragain,sowehaveatwodayupstreak,i.e.theStreakDurationvalueis2.
OnDay4,theclosingpricefalls,givingusaonedaydownstreak.TheStreakDurationvalueis
negative(1)becausethepricemovementisdown,notup.Thedownwardtrendcontinueson
This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |36

Days5and6,whichourStreakDurationreflectswithvaluesof2and3.OnDay7theclosing
priceisunchanged,sotheStreakDurationissetto0indicatingneitheranupclosenoradown
close.Finally,onDay8theclosingpricerisesagain,bringingtheStreakDurationbackto1.
ThesecondaspectofthesolutionistoapplytheRSIcalculationtothesetofStreakDuration
values.Bydefault,ConnorsRSIusesa2periodRSIforthispartofthecalculation,whichwe
denoteasRSI(Streak,2).Theresultisthatthelongeranupstreakcontinues,thecloserthe
RSI(Streak,2)valuewillbeto100.Conversely,thelongerthatadownstreakcontinues,the
closertheRSI(Streak,2)valuewillbeto0.Thus,wenowhavetwocomponentsRSI(Close,3)
andRSI(Streak,2)thatbothusethesame0100scaletoprovideaperspectiveonthe
overbought/oversoldstatusofthesecuritywereevaluating.
RelativeMagnitudeofPriceChange:ThefinalcomponentofConnorsRSIlooksatthesizeof
todayspricechangeinrelationtopreviouspricechanges.WedothisbyusingaPercentRank
calculation,whichmayalsobereferredtoasapercentile.Basically,thePercentRankvalue
tellsusthepercentageofvaluesinthelookbackperiodthatarelessthanthecurrentvalue.
Forthiscalculation,wemeasurepricechangenotindollarsandcents,butasapercentageof
thepreviousdaysprice.Thispercentagegainorlossistypicallyreferredtoastheoneday
return.Soifyesterdaysclosingpricewas$80.00,andtodayspriceis$81.60,theonedayreturn
is($81.60$80.00)/$80.00=0.02=2.0%.
TodeterminethePercentRank,weneedtoestablishalookbackperiod.ThePercentRank
valueisthenthenumberofvaluesinthelookbackperiodthatarelessthanthecurrentvalue,
dividedbythetotalnumberofvalues.Forexample,ifthelookbackperiodis20days,thenwe
wouldcomparetodays2.0%returntotheonedayreturnsfromeachoftheprevious20days.
Letsassumethatthreeofthosevaluesarelessthan2.0%.WewouldcalculatePercentRankas:
PercentRank=3/20=0.15=15%
ThedefaultPercentRanklookbackperiodusedforConnorsRSIis100,orPercentRank(100).We
arecomparingtodaysreturntotheprevious100returns,orabout5monthsofpricehistory.To
reiterate,largepositivereturnswillhaveaPercentRankcloserto100.Largenegativereturns
willhaveaPercentRankcloserto0.
ThefinalConnorsRSIcalculationsimplydeterminestheaverageofthethreecomponentvalues.Thus,
usingthedefaultinputparameterswouldgiveustheequation:

ConnorsRSI(3,2,100)=[RSI(Close,3)+RSI(Streak,2)+PercentRank(100)]/3

Theresultisaveryrobustindicatorthatismoreeffectivethananyofthethreecomponentsused
individually,andinmostcases,alsomoreeffectivethancombiningthethreecomponents
independently.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |37

ReceiveConnorsRSIReadingsforFreefromtheTradingMarketsScreener

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |38

TheS&P500LowVolatilityGrowthPortfolio

AreYouLookingForGrowthInBullMarkets&SafetyIn
BearMarketsFromYourInvestmentsWithS&P500
Stocks?
Introducing:TheS&P500LowVolatilityGrowthPortfolioGrowthinaBullMarket,SafetyinaBear
Market

Youdon'tneedanintroductiontoS&P500stocks.Butyoumayneedabetterwaytotradethemto
maximizeyourreturns.

ThemajorityofmoneyinretirementpensionandretirementaccountsisinS&P500stocks.Thisiswhere
thesafermoneylikestoinvest.

S&P500companiestendtobebiggerandsafercompanies.ManyareFortune500companieswith
worldwideoperationssellingproductstomillionsofcustomers.

Whenthestockmarketrises,youwanttoownsolidS&P500stocks.

Whenthestockmarketdeclines,youwanttobeincash.

ThisisexactlywhatTheS&P500LowVolatilityGrowthPortfoliodoesforyou.

TheS&P500LowVolatilityGrowthPortfoliodeliverstradingsignalsforS&P500stocks,providingyou
anoptimizedbalanceofbothgrowthandsafetyforyourcapital.

Inrisingmarketsyouarelongbluechipstocks.Indecliningmarketsyouareincash.

WouldyoulikethepotentialtoearnreturnslikethesefromS&P500stocks?
SamplesimulatedhistoricalbacktestresultsfromTheS&P500LowVolatilityGrowthPortfolio.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.

P a g e |39

YouAlreadyLearnedThisOverthePastDecade'BuyandHold'isDead!
Youunderstandthatmarketschangeastheyhaveoverthepastdecade.Inordertotakeadvantageof
changingmarketconditions,TheS&P500LowVolatilityGrowthPortfolioisconstantlyreviewed,
monitored,andupdatedbyConnorsResearch.
ThebacktestresultsindicatethatTheS&PLowVolatilityGrowthPortfolioconsistentlyoutperformsthe
S&P500on"buy&hold".
Sampleequitycurvefor$100KinvestedinTheS&P500LowVolatilityGrowthPortfoliovs.theS&P500.

This document cannot be reproduced without the expressed written permission of Connors Research, LLC.