HEDGEFUNDIE's Excellent Adventure (UPRO/TMF)

UPRO vs. TQQQ SkiptomaincontentSkiptosecondarymenuSkiptoprimarysidebarSkiptofooterHerewediveintothefamous“ExcellentAdventure”fromHedgefundieandhowtoimplementit.InterestedinmoreLazyPortfolios?Seethefulllisthere.Disclosure: Someofthelinksonthispagearereferrallinks.Atnoadditionalcosttoyou,ifyouchoosetomakeapurchaseorsignupforaserviceafterclickingthroughthoselinks,Imayreceiveasmallcommission.Thisallowsmetocontinueproducinghigh-quality,ad-freecontentonthissiteandpaysfortheoccasionalcupofcoffee.Ihavefirst-handexperiencewitheveryproductorserviceIrecommend,andIrecommendthembecauseIgenuinelybelievetheyareuseful,notbecauseofthecommissionIgetifyoudecidetopurchasethroughmylinks.Readmorehere.Inahurry?Herearethehighlights:HedgefundieiswasamemberoftheBogleheadsforum.HedgefundiecreatedathreadinFebruary2019proposinga3xleveragedETFinvestingstrategybasedonriskparityusingtheS&P500index(UPRO)andlong-termtreasurybonds(TMF)heldina40/60allocation.ThethreadlaterexpandedintoaPart2.Hedgefundielaterupdatedthestrategy’sassetallocationinAugust2019to55/45UPRO/TMF.Extensivebacktesting,discussion,andanalysiswithinthethreadbymembersoftheBogleheadsforumsupportsthevalidityandpotentialmarketoutperformanceofthestrategy.Theproposedstrategycallsforquarterlyrebalancing.Severaldifferentprotocols/variationsofthestrategyemergedintheExcellentAdventurethread,includingmonthlyrebalancing,rebalancingbands,andvolatilitytargetingwithvariouslookbackperiods.SomeusershaveaddedadashofTQQQ(3xtheNASDAQ100index)foraminortechtilt,asBigTechhashadastellarrunrecently.ItisrecommendedtoimplementthestrategywithinaRothIRAonM1Finance,toavoidtaximplicationsandtomakeregularrebalancingseamlessandeasy.Disclaimer: WhileIlovedivingintoinvesting-relateddataandplayingaroundwithbacktests,Iaminnowayacertifiedexpert.Ihavenoformalfinancialeducation.Iamnotafinancialadvisor,portfoliomanager,oraccountant.Thisisnotfinancialadvice,investingadvice,ortaxadvice.Theinformationonthiswebsiteisforinformationalandrecreationalpurposesonly.Investmentproductsdiscussed(ETFs,mutualfunds,etc.)areforillustrativepurposesonly.Itisnotarecommendationtobuy,sell,orotherwisetransactinanyoftheproductsmentioned.Doyourownduediligence.Pastperformancedoesnotguaranteefuturereturns.Readmylengthierdisclaimerhere.ContentsVideoWhoIsHedgefundie?WhatIstheHedgefundieStrategy?UPROvs.TQQQWhyNot100%UPRO?MyHedgefundieAdventurePerformanceAlternativeOptionsTotheHedgefundiePortfolioAddressingConcernsOverLong-TermTreasuryBondsReducingVolatilityandDrawdownsandHedgingAgainstInflationandRisingRatesTheHedgefundiePortfolioETFPieforM1Finance(UPRO/TMF)VideoPrefervideo?Watchithere:WhoIsHedgefundie?HedgefundieiswasamemberoftheBogleheadsforumwhocreatedanow-famousthreadontheforumproposinga3xleveragedETFstrategy.WhatIstheHedgefundieStrategy?TheHedgefundiestrategy–thewildrideofwhichisknownas“Hedgefundie’sExcellentAdventure”–isbasedonariskparityallocationofleveragedstocks(3xtheS&P500indexviaUPRO)andleveragedlong-termtreasurybonds(3xtheICEU.S.Treasury20+YearBondIndexviaTMF).Notethat“HFEA”istheshorthandinitialismforthenameofthestrategy,notthetickerforafund.Riskparityisaportfolioallocationstrategyinwhich,consistentwithModernPortfolioTheory(MPT),riskisspreadevenlyamongassetswithintheportfoliobylookingatthevolatilitycontributedbyeachasset,therebyattemptingtooptimizereturnsperunitofrisk(Sharpe).Iexplaineditmorehere.Paritybetweenstocksandlongtreasuriesisroughlyachievedat40/60.TheHedgefundiestrategyreliesheavilyonthenegativecorrelation(oratleast,uncorrelation)betweenstocksandlong-termtreasurybonds,whereinthebondsprovideabufferduringstockdrawdowns.Long-termtreasuriesarechosenpreciselybecausetheyaremorevolatilethanshorter-durationbondsandbecauseoftheirdegreeofnegativecorrelationtostocks,inordertosufficientlycounteractthedownwardmovementofa3xleveragedstockspositioninacrash.Idelvedintothesespecificbenefitsoftreasurybondshere.Thisconceptisbasedonthesimplehistoricalprincipleofimprovingrisk-adjustedreturn(Sharpe)overlongperiodsbyholdinguncorrelatedassets,suchasatraditional60/40stocks/bondsportfolio,asopposedto100%stocks.Inanutshell,thisisawaytoholdUPROlongterminamuchmoresensibleway.ConsistentwiththeideaofLifecycleInvesting,thisheavily-leveragedstrategyisbettersuitedforyounginvestorswithalongtimehorizonwhocanaffordtoberiskyearlyintheirinvestmenthorizon.Hedgefundieadvocatesfortreatingthisstrategylikea“lotteryticket”andnotusingitwithasignificantportionofyourtotalportfoliovalue.Criticsandnaysayersreflexivelyexclaimtheoft-cited,overblown,platitudinous“LeveragedETF’saren’tmeanttobeheldlong-termbecauseofvolatilitydecay,”but,inshort,thatdoesn’tconcernme.Moreover,thatsamevolatilitydecaycanactuallyhelpwhenupwardmovementwithpositivemomentumisoccurring.Iwouldalsoarguethataslongasyoucanstomachthevolatility,amajordropshould[eventually]befollowedbyamajorrebound;3xhurtsonthewaydownbuthelpsonthewayup.UPROfromProSharesandTMFfromDirexionwerechosenduetotheirlowtrackingerrorandhighvolume;again,we’regetting300%exposuretotheS&P500andlong-termtreasurybonds,respectively.Theproposedstrategycallsforquarterlyrebalancing.Severaldifferentprotocols/variationsofthestrategyemergedastheExcellentAdventurethreadprogressed,includingmonthlyrebalancing,rebalancingusingbands,andvolatilitytargetingwithvariouslookbackperiods.I’dkeepitsimpleandavoidcheckingitoften;Icanseeitbeingveryeasytogetemotionalwiththisstrategyandabandonyourplan.ItisrecommendedtoimplementthestrategywithinaRothIRAonM1Finance,toavoidtaximplicationsandtomakeregularrebalancingseamlessandeasy.IknowthissoundssaIespitchy,butifyou’rewantingtousethisstrategyinataxableaccount,IwouldargueitmakesevenmoresensetouseM1Financebecauseifyou’rechoosingtoputinnewdeposits,thesystemwillautomaticallyrebalancetheportfolioforyoubydirectingnewdepositstobuytheunderweightasset,therebyallowingyoutoavoidcapitalgainstaxesthatwouldotherwisebeincurredwithamanualrebalance.Thisismoreimpactfulthanitmightsoundatfirst.Theseare3xleveragedETFs;theycanveryquicklygetoutofbalance.Forexample,let’ssayyoustartoutattheprescribed55/45andstockstakeoffandbondssuffer,whichcausesittostrayto75/25afteronlyamonth.Notgood.Atthispointyou’dhavetoincurshorttermcapitalgainstaxes(ouch!)justtogetthingsbackinbalance.Granted,atacertainpoint,yournewdepositsmaynotbesufficientlylargeenoughtoprovidethefullrebalancingeffectontheirown,butthatwouldbeagreatproblemtohave.Utilizingatraditional,unleveraged40/60stocks/bondsportfolio,comparedtoanall-equitiesportfolio,hasrelativelylowvolatilityandshouldproducehigherrisk-adjustedreturn(Sharpe)overlongtimeperiods,butwouldalsolikelyunderperformanall-equitiesportfoliointermsoftotalreturn.Thesolution,Hedgefundiemaintains,isapplyingleverage.We’reattemptingtoacceptariskprofilesimilartothatoftheS&P500,butwithmuchhigherexpectedreturns.Hedgefundieupdatedtheapproach6monthsafterpostingtheoriginalstrategy,optingtomovetoa55/45UPRO/TMFallocationfromtheprevious40/60riskparityallocation.Hedgefundie’sreasonsarelaidouthere,basedprimarilyonthepremisethatthestocksportionofthestrategyistheprimarydriverofthestrategy’sreturnsandthatthemainpurposeofholdingthetreasurybondsisessentiallyas“insurance”incaseofastockmarketcrash.Intrinsically,we’rerelyingonUSstocksandlong-termtreasuriesnotcrashingintandem.Atthetimeofwriting,theseassetshaveaseeminglyreliablynegativecorrelationcloseto-0.5onaverage.AkeyfundamentalassumptionofthisstrategythatHedgefundieproposesisthattheUSwillnotreturntopre-Volcker(pre-1982)monetarypolicy.Thatis,we’llbeabletosignificantlymitigateoraltogetheravoidrunawayinflationperiodslikethelate1970’s,duringwhichtimebondssufferedgreatly.Stocksandlong-termtreasurybondsdonothaveaperfect-1correlation.Sometimestheymoveinthesamedirection.Thisisactuallyagoodthing.Historically,whentheseassetsmovedinthesamedirection,itwasusuallyup.Ondayswhenstocksdropped,long-termtreasuriesfairlyreliablyrosesignificantlytomitigatethetotalloss.Simulatedreturnsgoingbackto1987looklikethis:Herearetherollingreturns:Belowarethedrawdowns.NoticethesmallerdrawdownsinmostcasescomparedtotheS&P500:IagreewithHedgefundie’sassertionthatextremelyvolatileassetslikegold,commodities,smallcaps,etc.wouldsufferworsefromvolatilitydecayandwouldnotimprovethestrategy’sdiversificationandreturn.Internationaldevelopedmarketsmaybeaviableoptiontoinclude,butBogleheadmembersiamondfoundissueswiththeDZKETF,whichendedupclosinginOctober,2020anyway.Ifyouwantedtoforsomereason,youcouldalsousetheslightlymoreexpensiveSPXLinsteadofUPRO.Theirliquidityandperformanceshouldbenearlyidentical.Makenomistakethatthisisariskystrategybyitsverynature.ReaduponleverageandthenatureofleveragedETF’sbeforeemployingthisstrategy.Donotputyourentireportfoliointhisstrategy.Readmoredetailsandnuancesofthestrategyontheoriginalthreadhere.Ifyou’vegotthetime,there’salotoflearningtobehadthroughouttheentirethread.ThethreadhasexpandedintoaPart2here.UPROvs.TQQQSomeusershaveaddedadashofTQQQ(3xtheNASDAQ100index)foraminortechtilt,asBigTechhashadastellarrunrecently.OthersstillareusingTQQQastheentireequitiespositionfortheHFEAstrategy.Ipersonallythinkthisisunnecessaryandispurelyperformancechasingasaproductofrecencybias.ImagineforasecondthatthisisJanuary,2010.Afterthepreviousdecade,theS&P500isdownbyabout10%forthattimeperiodversustheNasdaq100beingdownabout50%.WouldyoustillbeasenthusedaboutTQQQ?Logically,weshouldbe more willingtobuywhenpricesarelow,butI’dbewillingtobetthehonestanswertothisquestionformostfolkswouldbe“no.”Arationalinvestorshouldwantto avoid expensivestocksandbuycheapstocks,butthisunfortunatelyisn’thowinvestors’highly-emotionalbrainswork.TQQQhasbeatenUPROhistoricallyintermsofsheerperformance.Butdon’tsuccumbto recencybias.Pastperformancedoesnotindicatefutureperformance.Moreimportantly,largecapgrowthstocksarenowlookingextremelyexpensiverelativetohistoryandareatthevaluationswesawin2000attheheightofthetechbubble,meaningtheynowhavelowerfutureexpectedreturns.Tomakethingsworse,fundamentalsofthesecompaniesdonotexplainthesevaluations.Thecurrentsituationissimplytheresultofanexpansionofpricemultiples.Valuestocks,ontheotherhands,arelookingextremelycheap,meaningtheynowhavegreaterexpectedreturns.Ofcourse,weexpectValuetooutperformeverydaywhenwewakeupanywayduetowhatwethinkisa riskfactorpremium.IfyoubuyTQQQ,youwon’townanyValuestocks.TQQQispurelylargecapgrowthstocks,thesegmentwith lower expectedreturns.Youalsowon’townanysmall-ormid-capstocks,whichhaveoutperformedlargestockshistorically.ThevaluationspreadbetweenValueandGrowthwasrecentlyaslargeasit’severbeen.Historically,widevaluespreads havealsoreliablypreceded massiveoutperformancebyValue.Attheendoftheday,we’restillpayingforadiscountedsumofallfuturecashflows;Growthcannotgetmoreexpensiveforever.Unfortunately,there’snoleveragedValueETF.Peopleliketoclaim“techisthefuture!”Thatmaybetrue,butthatdoesn’thavemuchtodowithstockmarketreturns,whicharenotcorrelatedwithGDP.Theeconomyisnotthestockmarket,andthestockmarketisnottheeconomy.Rememberthatextremelyhighexpectationsforthesetechfirmsare alreadypricedin,andtheywillhaveto exceed thoseexpectationsinordertobeatthemarket.Moreover,goodcompaniestendtomakebadstocksandbadcompaniestendtomakegoodstocks.Alsorememberthatyoudon’tneeda“techtilt”anyway;themarketisalreadyover30%techatthispoint.TheNASDAQ100isbasicallyatechindexatthispoint;it’srealisticallyabout70%tech,posingasectorconcentrationrisk,whichisuncompensatedrisk.WhileIdon’temployorcondonemarkettiming,wealsomustacknowledgethefactthatwemayseerisinginterestratessometimeinthenearfuture,andTQQQinherentlyhasmoreinterestrateriskthanUPRO.Moreover,TQQQbydefinitionexcludesFinancials,whichtendtodowellwheninterestratesrise.Nowmaybetheworsttimetooverweightlargecapgrowth,butmytimemachineisbroken.Onlytimewilltellwhichindexoutperforms.Wecan’tknowthefuture,butIwouldarguethat’sthereasonforbroad diversification inthefirstplace.WhyNot100%UPRO?Ifwe’reexpectingUPROtobethedriverofthestrategy’sreturns,whynotgo100%UPRO?HedgefundieaddressedthisintheoriginalBogleheadsthreadbypointingoutthatindoingso,we’dprobablyexpectsuperdeepdrawdownsfromwhichitmaytakedecadestorecover.Here’sabacktestshowing40/60UPRO/TMF(Portfolio1)vs.100%UPRO(Portfolio2)toillustrate:Source:PortfolioVisualizer.comHere’sUPROvs.theS&P500goingbackto1955:Source:PortfolioVisualizer.comMyHedgefundieAdventurePerformanceTrackingthequarterlychangeinperformance(relativetotheinitialvalue;nonewdeposits)ofmyHedgefundieAdventureinmyownportfoliostartingOctober1,2019:01/01/2020:+7%04/01/2020:-2%07/01/2020:+35%10/01/2020:+54%01/01/2021:+79%04/01/2021:+67%07/01/2021:+105%10/01/2021:+105%01/01/2022:+150%04/01/2022:+98%AlternativeOptionsTotheHedgefundiePortfolioIfyouwanttoutilizealeveragedstrategysimilartothatproposedbyHedgefundiebutbecompletelyhandsoff,PIMCOhasbeendoingsomethingsimilarforyearswiththeirStocksPLUSLongDurationFund(PSLDX)since2007.Ireviewedthefundhere.Notethatyoucanonlyaccessthisfundthroughcertainbrokers,anditmayhaveaminimuminvestmentrequirementandtransactionfees.Thosedetailsarebeyondthescopeofthispost;askyourbrokerifit’savailabletoyou.Similarly,ifyou’redoingthiswithasmallportionofyourportfolioorifyouwanttoemployaleveragedstrategyinataxableaccount,WisdomTree’sNTSXmaybeasuitableoption,effectivelyproviding1.5xleverageonatraditional60/40stocks/bondsportfolio.Itholds90%straightS&P500stocksand10%treasuryfuturestoachieveeffectivenotionalexposureof90/60stocks/bonds.Ireviewedthefundhere.BogleheadsuserMotoTrojanproposedavariantbywhichyoucanmatchthevolatilityofHedgefundie’s55/45UPRO/TMF,tonedowntheleverageabit,andsavesomeontheexpenseratioofTMFbyutilizingVanguard’sExtendedDurationTreasuryETF(EDV)inaratioof43/57UPRO/EDV.Here’sanM1pieforthat.Thisvariantwouldalsobemoretax-efficientthantheoriginalstrategythatusesTMFifyou’redoingthisintaxable.Rapidlyrisinginterestratesand/orrunawayinflationaretheprimaryrisksforthisstrategy.Ifthoseconcernsarematerialtoyouandmakeyouhesitantaboutthisstrategy,orifyousimplywantmorediversificationacrossassettypes,thenaleveragedAllWeatherPortfoliomayappealtoyou.Therearealsosomediversifierslistedbelow.AddressingConcernsOverLong-TermTreasuryBondsI’vegottenalotofquestionsabout–andalotofthediscussionintheoriginalBogleheadsthreadhasbeenabout–theuse,utility,andviabilityoflong-termtreasurybondsasasignificantchunkofthisstrategy.I’llbrieflyaddressandhopefullyquelltheseconcernsbelow.Again,bydiversifyingacrossuncorrelatedassets,wemeanholdingdifferentassetsthatwillperformwellatdifferenttimes.Forexample,whenstockszig,bondstendtozag.Those2assetsareuncorrelated.Holdingbothprovidesasmootherride,reducingportfoliovolatility(variabilityofreturn)andrisk.Commoncommentsnowadaysaboutbondsinclude:“Bondsareuselessatlowyields!”“Bondsareforoldpeople!”“Longbondsaretoovolatileandtoosusceptibletointerestraterisk!”“Corporatebondspaymore!”“Interestratescanonlygoupfromhere!Bondswillbetoast!”“Bondsreturnlessthanstocks!”Sowhylongtermtreasuries?Itisfundamentallyincorrecttosaythatbondsmustnecessarilylosemoneyinarisingrateenvironment.Bondsonlysufferfromrisinginterestrateswhenthoseratesarerisingfasterthanexpected.Bondshandlelowandslowrateincreasesjustfine;lookattheperiodofrisinginterestratesbetween1940andabout1975,wherebondskeptrollingattheirparandpaidthatsweet,steadycoupon.Ratesalsorosesteadilyfrom2016tomid-2019,duringwhichtimeTMFdeliveredapositivereturn.Bondpricingdoesnothappeninavacuum.Herearesomemoreexamplesofperiodsofrisinginterestrateswherelongbondsdeliveredapositivereturn:From1992-2000,interestratesrosebyabout3%andlongtreasurybondsreturnedabout9%annualizedfortheperiod.From2003-2007,interestratesrosebyabout4%andlongtreasurybondsreturnedabout5%annualizedfortheperiod.From2015-2019,interestratesrosebyabout2%andlongtreasurybondsreturnedabout5%annualizedfortheperiod.Newbondsboughtbyabondindexfundinarisingrateenvironmentwillbeboughtatthehigherrate,whileoldonesatthepreviouslowerratearesoldoff.You’renotstuckwiththesameyieldforyourentireinvestinghorizon.Weknowthattreasurybondsareanobjectivelysuperiordiversifieralongsidestockscomparedtocorporatebonds.ThisisalsowhyIdon’tusethepopulartotalbondmarketfundBND.Ithasbeennotedthatthisgreaterdegreeofuncorrelationbetweentreasurybondsandstocksisconvenientlyamplifiedduringperiodsofmarketturmoil,whichresearchersreferredtoascrisisalpha.Again,rememberweneedandwantthegreatervolatilityoflong-termbondssothattheycanmoreeffectivelycounteractthedownwardmovementofstocks,whichareriskierandmorevolatilethanbonds.We’reusingthemtoreducetheportfolio’svolatilityandrisk.Morevolatileassetsmakebetterdiversifiers.Mostoftheportfolio’sriskisstillbeingcontributedbystocks.Let’suseasimplisticriskparityexampletoillustrate.RiskparityforUPROandTMFisabout40/60.Ifwewanttoslidedownthedurationscale,we mustnecessarily decreaseUPRO’sallocation,asweonlyhave100%ofspacetoworkwith.RiskparityforUPROandTYD(orEDV)isabout25/75.ParityforUPROandTLTisabout20/80.etc.Simplykeepingthesame55/45allocation(forHFEA,atleast)andswappingoutTMFforashorterdurationbondfunddoesn’treallysolveanythingforus.ThisiswhyI’vesaidthatwhileit’snotperfect,TMFseemstobethe“leastbad”optionwehave,aswecan’tleverintermediates(TYD)past3xwithouttheuseoffutures.Thisone’sprobablythemostimportant.We’renottalkingaboutbondsheldinisolation,whichwouldprobablybeabadinvestmentrightnow.We’retalkingabouttheminthecontextofadiversifiedportfolioalongsidestocks,forwhichtheyarestilltheusualflight-to-safetyassetduringstockdownturns.Specifically,forthisstrategy,thepurposeofthebondssideispurelyasaninsuranceparachuteintheeventofastockcrash.Thoughtheyprovidedamajorboosttothisstrategy’sreturnsoverthelast40yearswhileinterestratesweredropping,we’renotreallyexpectinganyrealreturnsfromthebondssidegoingforward,andwe’reintrinsicallyassumingthatthestockssideistheprimarydriverofthestrategy’sreturns.Evenifrisingratesmeanbondsareacomparativelyworsediversifier(forstocks)intermsoffutureexpectedreturnsduringthatperioddoesnotmeantheyarenotstillthebestdiversifiertouse.Similarly,short-termdecreasesinbondpricesdonotmeanthebondsarenotstilldoingtheirjobofbufferingstockdownturns.Historically,whentreasurybondsmovedinthesamedirectionasstocks,itwasusuallyup.Interestratesarelikelytostaylowforawhile.Also,there’snoreasontoexpectinterestratestorisejustbecausetheyarelow.Peoplehavebeenclaiming“ratescanonlygoup”forthepast20yearsorsoandtheyhaven’t.Theyhavegraduallydeclinedforthelast700yearswithoutreversiontothemean.Negativeratesaren’toutofthequestion,andwe’reseeingthemusedinsomeforeigncountries.Bondconvexitymeanstheirasymmetricrisk/returnprofilefavorstheupside.Again,Iacknowledgethatpost-Volckermonetarypolicy,resultinginfallinginterestrates,hasdriventheparticularlystellarreturnsoftheragingbondbullmarketsince1982,butIalsothinktheFedandU.S.monetarypolicyarefundamentallydifferentsincetheVolckerera,likelyallowingustoaltogetheravoidrunawayinflationenvironmentslikethelate1970’sgoingforward.Bondpricesalreadyhaveexpectedinflationbakedin.DavidSwensensummeditupnicelyinhisbookUnconventionalSuccess:“Thepurityofnoncallable,long-term,default-freetreasurybondsprovidesthemostpowerfuldiversificationtoinvestorportfolios.”Ok,bondsrantover.Ifyoustillfeelsomedissonance,thenextsectionmayoffersomesolutions.ReducingVolatilityandDrawdownsandHedgingAgainstInflationandRisingRatesIt’sunlikelythatanyofthefollowingwillimprovethetotalreturnoftheportfolio,andwhetherornotthey’llimproverisk-adjustedreturnisupfordebate,butthoseconcernedaboutinflation,risingrates,volatility,drawdowns,etc.,and/orTMF’sfutureabilitytoadequatelyserveasaninsuranceparachute,maywanttodiversifyabitwithsomeofthefollowingoptions:LTPZ–longtermTIPS–inflation-linkedbonds.FAS–3xfinancials–bankstendtodowellwheninterestratesrise.EDC–3xemergingmarkets–diversifyoutsidetheU.S.UTSL–3xutilities–lowestcorrelationtothemarketofanysector;tendtofarewellduringrecessionsandcrashes.YINN–3xChina–lowlycorrelatedtotheU.S.UGL–2xgold–usuallylowlycorrelatedtobothstocksandbonds,butalong-termexpectedrealreturnofzero;no3xgoldfundsavailable.DRN–3xREITs–arguablediversificationbenefitfrom“realassets.”EDV–U.S.TreasurySTRIPS.TYD–3xintermediatetreasuries–lessinterestraterisk.UDOW–3xtheDow–greaterloadingonValueandProfitabilityfactorsthanUPRO.TNA–3xRussell2000–smallcapsfortheSizefactor.TAIL–OTMputoptionsladdertohedgetailrisk.MostlyintermediatetreasurybondsandTIPS.TheHedgefundiePortfolioETFPieforM1Finance(UPRO/TMF)Again,mostusersareutilizingM1FinancetodeploytheHedgefundiestrategyduetoitsdynamicrebalancingwithnewdeposits,zerotransactionfees,anditssimple,1-clickrebalancethatyoucandoquarterly.Ittakesnomorethan30secondsevery3months.IwroteacomprehensivereviewofM1Financehere.Theriskparity40/60portfoliowouldbethispiewhichlookslikethis:40%UPRO60%TMFToaddthispietoyourportfolioonM1Finance,justclickthislinkandthenclick“Savetomyaccount.”Theupdated55/45portfoliowouldbethispiewhichlookslikethis:55%UPRO45%TMFToaddthispietoyourportfolioonM1Finance,justclickthislinkandthenclick“Savetomyaccount.”CanadianscanfindtheaboveETFsonQuestradeorInteractiveBrokers.InvestorsoutsideNorthAmericacanuseeToroorpossiblyInteractiveBrokers.M1Financecurrentlyhasapromotionforupto$250wheninitiallyfundinganinvestmentaccount,asoutlinedbelow:Theyalsocurrentlyhaveatransferbonuspromotionforupto$2,500whentransferringanexistingaccountfromanotherbrokerage,asoutlinedbelow:LearnMoreDisclosures:IamlongPSLDX,NTSX,UPRO,andTMFinmyownportfolio.InterestedinmoreLazyPortfolios?Seethefulllisthere.Disclaimer: WhileIlovedivingintoinvesting-relateddataandplayingaroundwithbacktests,Iaminnowayacertifiedexpert.Ihavenoformalfinancialeducation.Iamnotafinancialadvisor,portfoliomanager,oraccountant.Thisisnotfinancialadvice,investingadvice,ortaxadvice.Theinformationonthiswebsiteisforinformationalandrecreationalpurposesonly.Investmentproductsdiscussed(ETFs,mutualfunds,etc.)areforillustrativepurposesonly.Itisnotarecommendationtobuy,sell,orotherwisetransactinanyoftheproductsmentioned.Doyourownduediligence.Pastperformancedoesnotguaranteefuturereturns.Readmylengthierdisclaimerhere.RelatedPostsSecondGrader’sStarterPortfolioReviewandETFPieThe3BestCorporateBondETFs(2FromVanguard)HowToBuyAppleStockWith$100–HowToInvestinAppleThe5BestConsumerStaplesETFsDayTradingvs.InvestinginStocks–WhichIsBetter?AboutJohnWilliamsonAnalyticalandentrepreneurial-mindeddatanerd,usabilityenthusiast,Boglehead,andOxfordcommaadvocate.IleadthePaidSearchmarketingeffortsatGildGroup.I'mnotabigfanofsocialmedia,butyoucanfindmeonLinkedInandReddit.ReaderInteractionsCommentsIknowyounevertimethemarket,butisthereanyreasontothinkrightnowisaparticularlygoodorbadtimetostartthisstrategy?Doesleveragehavedifferentrules?ReplyImpossibletosay.Besttimetoplantatreewas20yearsago;secondbesttimeistoday.ReplyGreatwebsiteandarticle.Hastherebeenanyback-testingonthreshold-band-basedrebalancinginsteadofquarterly.Rebalanceat10or15,i.e.,ifwestartat55/45,rebalanceat65/35or45/55.ReplyHowwouldyoucomparethethreewaysofgoingaboutthisstrategy?Isthereareasonyouutilizeallthree?UPRO/TMFvs.PDLDXvs.NTSX?IsUPROroutethemostrisk/rewardandNTSXtheleast?ReplyTaxefficiencyandsimplicity.ReplyItseemsthecurrentenvironmentistheworst-casescenarioforthisstrategy:rapidlyrisingrates,fallingequityprices,andrisinginflation.Anyupdateonitsperformance?ReplyYes,thiswasalwaystheachillesheelofthisstrategy.Quarterisn’toveryet.Givemeafewdaystoupdateperformance.I’mstillupnearly100%fromlate2019.ReplyDoIhaveanydownsideindoingthisintheTraditionalIRAvsRothIRA?RebalancingTraditionalIRAwouldn’ttriggerataxeventcorrect?I’mnewtoUSretirement.Muchappreciated!ReplyCorrect,Mark.FineforTraditionalorRothIRA.ReplyItlookslikeTMFisnotprovidingmuchnon-correlationlately.Wouldn’t40-45%cashbeabetterholdthanTMF?ReplyImpossibletoknowthefuture.Holdingcasheffectivelyjustmeansyou’retakingonlessequitiesleverage.We’reseeinghighinflationinthefaceofrisinginterestrates.Thiswasalwaystheachillesheelofthisstrategy,andwasdiscussedatlengthintheoriginalthread.Nottomentionawarhappeningatthesametime.Thisisalong-termstrategyofnecessity.ReplyAwesomearticle,thanksforthewriteup.Howwouldthe55/45splitcomparetothe40/60splitand100%UPROinthe“whynot100%UPRO”segment?ReplyWhoopsnevermindthelastpartofmyquestion–quarterly!ReplyHiJohn,Thanksforyourgreatinsights!I’mwonderingwhatyourthoughtsareaboutsubstitutingSPXUforTMFinsomeproportionsinceitwouldbeevenmoreconsistentlyinverselycorrelatedwithUPRO.Also,howfrequentlydoyourecommendrebalancingHedgefundie?ReplyJack,SPXUisjusttheinverseofUPRO.Thesumoftheirreturnswouldbezero.Wedon’twantaperfect-1correlation.ReplyHI,howtoperformquarterlyrebalancingifyourincomedoesn’tallowyoucontributetorothiraaccount?Iamworriedquarterlyrebalancingwouldleadtoheavytaximplications.ReplyDoabackdoorRothifyoucan.Intaxable,ifyouareregularlydepositing,youcanbuytheunderweightassetwithnewdeposits(automaticifusingM1)untilthepointatwhichdepositsarenotsufficientlylargeenoughtorebalance.Thestrategyisnotmeanttobeusedinataxableaccount.ReplyThanksabunchforthis!NowI’mgoingtogoandhoponM1andgetleveredtothe****!ReplyThisisawonderfulpresentationofafascinatingstrategy.GiventhecurrentmarketconditionswithFEDclearlytiltingtowardsalessaccommodativestance,doyoustillthinkHFEAwillbeagoodlongtermstrategytoimplement?Inanenvironmentwithinterestratehikesand-mostimportantly-theFEDoffloadingtheirbalancesheet,willthisstrategybeabletocontinuetooutperformtheindexes?ReplyThanks!Impossibletosay.ReplyHellofromtheUK,Iloveyourwebsitebeenlookingforsomethinglikethisforalong,ratherthanthetrialanderrorIhavebeendoingforalongtime.DoyouknowofanyUK(GBP)tradedETF’sIcanuseplease.Ican’tuseUSETF’sbecausethecurrencyexchangefeethatbrokerschargeintheUKareskyhigh.Thanks,DReplyTheinvestingpostsonthiswebsitehavebeenextremelyhelpfultome;thankyou!Iamwonderingwhatyourthoughtsareonthis55/45HFEAvsSSO?Wouldtheirriskbeconsideredsimilar?Whichdoyouthinkwouldhaveagreaterreturnwhenbacktesting?Yourpostondiversificationwasveryinformative;beforereadingthepost,IwouldsaythatIwasdefinitelyveryUSbiased.Couldonediversify55/45HFEAtoalsoincludeex-USstocks?HaveyouconsideredbacktestingyourmoreaggressiveGingerAlewith55/45HFEAreplacingVOO?55/45HFEA–30%AVUV–30%VEA–10%AVDV–10%VWO–10%DGS–10%ReplyThanks!I’dbemorelikelytogoHFEAthanSSO.Wecanshowempiricallythattheformerismoreefficient,whichwe’dexpect.SSOhasmoreriskbybeingasingleasset.Wegetabigdiversificationbenefitwhenweincludeuncorrelatedassets.Toansweryourperformancequestion,HFEAhashadgreatergeneralandrisk-adjustedreturnshistoricallythan200%equities.CanwediversifyHFEAwithex-USstocks?Yes,buttheliquidityisn’tgreat.InthesectionondiversifiersImentionedEDCandEURL.HaveIconsideredincludingHFEAintheGingerAleasareplacementforVOO?No.Thatwouldn’tmakemuchsense.ReplyJohn,greatarticle!Thanks!DoyouthinkNTSXat2xmarginwouldbemoreefficientthanUPRO/TMFinataxableaccounttoreducetaxesandfees?ReplyThanks,Eugene!Inarguablyyesitwould,butIdon’tthinkanybrokerisgoingtoletyouleveritupthatmuch,soit’sprobablyjustatheoreticalidea.ReplySeeingTMFdropping7.53%today.Aswellashavingmuchmoreinformationonratehikescomparedtowhenthestrategywascreated.Doyouthinkweshouldmakeanyadjustmenttothe60/40ratioormaybeevendropTMFaltogetherinfearofthecomingratehikes?Iamscared.ReplyImeant55/45sorry.ReplyNo.Ifyou’rehavingthesethoughtsfromasingledayof<10%movement,thisstrategyisnotforyou.ReplyJohn,Fantasticblogandadvice.Trulyappreciated.Iamcuriousforyourperspectiveondollar-costaveraging.Wouldyourecommendinvesting$10,00055/45andrebalancingquarterly,or$833permonthandrebalancingmonthly?Thankyou.ReplyImpossibletosay.Here’smypostonDCA.ReplyGreatinfohere,thanks!Iput25%ofmyportfolioinHedgeFundieinearlyOct.,butIchoseTQQQoverUPRO.“Plastics”willalwayslead…It’svolatilebutontrackforanice20%upquarter.🙂Q:Whynotsimplifyfurtherandgo50/50withHedgeFundie?Why55/45?Has50/50beenbacktested?Thanks!Reply55/45wasoptimalhistoricallyIIRC.TQQQisnotHedgefundie;itwasdiscussedatlengthintheoriginalthread.ReplyIrecall55%waschoseninordertoallowthe55tocarrythereturnsoftheportfolioevenifthe45wasjustincash.Reply45isnotincashforHFEA.ReplyLovethearticleJohn–acoupleofquestions.1)Itlooksasthoughthedrawdowninthisunderstatedasitonlycalculatesthemonth-enddrawdowns.In2020theactualdrawdowninMarchofthatyearreached~44%.Areyouawareofanysitethatcancalculatethein-monthdrawdownswhendoingbacktesting(lookslikepotfoliovisualizerdoesonlymonth-end)2)AreyouawareofanybrokersinCanadathatdoautomaticrebalancingwithoutanyintervention?Thanksandkeepupthegreatwork!JamesReplyThanks,James.1.Idon’toffthetopofmyhead.2.I’mnot.PossiblyWealthsimple?ReplyLookingforwardtotrythisstrategywithmynextRothIRAcontributionassumingthatIcanstillusethebackdoormethod.IstilldontunderstandwhyitsnotadvocatedtobuytheglobalmarketinsteadofusingUPROonly.Iwasthinkingofdoingthis:55%3xStocks45%3x20+USTreasuryBonds(ASHFdoesit)Inthat55%stocks,Iamthinkingofdoing70%US30%INT70%UPRO15%EDC15%EURLIknowUPROandEURLwillbecorrelatedbutwedon’tknowwhichonewillperformbetterandsoIwouldratherhaveboth.IunderstandtheliquidityissuesandthereisachanceEURLmightclosedownbutisthereanyotherdownsidesofhavingEURLinyourportfolio?Reply«OlderCommentsLeaveaReplyCancelreplyYouremailaddresswillnotbepublished.Requiredfieldsaremarked*CommentName*Email*Savemyname,email,andwebsiteinthisbrowserforthenexttimeIcomment.Don'tsubscribeAllRepliestomycommentsNotifymeoffollowupcommentsviae-mail.Youcanalsosubscribewithoutcommenting. ΔPrimarySidebarMostPopularRecentPostsViewAll...reportthisadFooterEmailFacebookInstagramLinkedInPinterestRSSTwitterYouTubeRedditAmazonAffiliateDisclosureOptimizedPortfolio.comisaparticipantintheAmazonServicesLLCAssociatesProgram,anaffiliateadvertisingprogramdesignedtoprovideameansforsitestoearnadvertisingfeesbyadvertisingandlinkingtoAmazon.com.EmailNewsletterSignuptoreceiveemailupdateswhenanewpostispublished. Don'tworry,Ihatespamtoo.Noads.Thiswebsiteusescookiestoimproveyourexperience.We'llassumeyou'reokwiththis,butyoucanopt-outifyouwish.Ok,GotItRejectReadMorePrivacy&CookiesPolicy ClosePrivacyOverviewThiswebsiteusescookiestoimproveyourexperiencewhileyounavigatethroughthewebsite.Outofthese,thecookiesthatarecategorizedasnecessaryarestoredonyourbrowserastheyareessentialfortheworkingofbasicfunctionalitiesofthewebsite.Wealsousethird-partycookiesthathelpusanalyzeandunderstandhowyouusethiswebsite.Thesecookieswillbestoredinyourbrowseronlywithyourconsent.Youalsohavetheoptiontoopt-outofthesecookies.Butoptingoutofsomeofthesecookiesmayaffectyourbrowsingexperience.Necessary NecessaryAlwaysEnabledNecessarycookiesareabsolutelyessentialforthewebsitetofunctionproperly.Thiscategoryonlyincludescookiesthatensuresbasicfunctionalitiesandsecurityfeaturesofthewebsite.Thesecookiesdonotstoreanypersonalinformation.Non-necessary Non-necessaryAnycookiesthatmaynotbeparticularlynecessaryforthewebsitetofunctionandisusedspecificallytocollectuserpersonaldataviaanalytics,ads,otherembeddedcontentsaretermedasnon-necessarycookies.Itismandatorytoprocureuserconsentpriortorunningthesecookiesonyourwebsite.SAVE&ACCEPT