@thunderhead1 This may become a somewhat lengthy post...but anyway. Short answer: I did buy heavily from March-May (I think it was about 20% cost base position at its highest), and I sold a bit when I started getting a bit nervous about the size (when it was ~45%). Just about any stock that I hold for >6 months, I hold for many years (I tend to cut my losses over 6-9 months on most positions, assuming I am wrong unless I have a great deal of confidence otherwise), although I will say I have traded a lot more than is typical for me with the volatility of the pandemic.
Long answer: 1. Regarding this specific position: - I have previously owned ADH (it was one of the many mistakes I made trying to invest in retail from 2016-2017). My mistakes when analysing retail were: (a) I didn't understand the point in the cycle (b) I didn't take enough of an outside view (c) I overpaid (d) I focussed on unimportant stuff that everyone says you are supposed to do like look at stock turn, days inventory outstanding etc as part of some misguided analytical process. (e) I realised the retail malaise was a global phenomenon, not just a local phenomenon, then surmised that if I wanted to invest in retail I should actually go after the beaten up more world famous brands. This assumption was made at the bottom of the cycle (when Powell raised rates)... - the one thing I did correctly was figure out what I really did wrong, which was, as it turns out, what everyone was doing: --> assuming that the current situation will last forever --> across the retail landscape, the debt-burdened brands were going into VA, and the stronger specialty stores were closing stores putting pressure on landlords, whilst all at the same time expanding the online offerings. This was not particularly insightful- just about very retailer was saying the exact same thing. The market realised before I did, and I made a note of the errors I had initially made. - regarding ADH specifically- I knew the business very well and I shopped for the first time in awhile in December 2019 and promptly was upsold on linen lovers membership. I immediately realised how powerful a paid membership program was, and was dismayed to see that this was just after a share price rise from the Mocka acquisition! Nevertheless, it was on my high priority hit list to buy on any weakness. - during March I honestly had no idea what the market was thinking- the price moves were insane and made absolutely no sense. The only thing I was confident of, was that the market was not accurately discounting the duration of disrupted economic activity- I did my own modelling using an exponential decay bespoke model that accurately (enough) predicted sharp downswings in case numbers. - I made my first big move in APT (I increased my position something like 100% at very close to the bottom) and it rose quite quickly. Thereafter I became pretty confident that I was sane and the market was insane, and gave me the confidence to back my assertions. Retail was high on my list and omnichannel was highest, as I knew retail would likely be disrupted the least and hospitality/travel the most- you can read my posts on ADH from March-May-> in summary there was actually enough publicly well known information to take pretty big positions at very very low prices.
2. Regarding position sizing in general I have my own methods of sizing that I have optimised for myself over a number of years now. (I had started writing a thread on it called "portfolio allocation" if you want to have a look at some of the numbers, which I stopped because noone else said anything that I found useful). There are some basic mathematical principles that I use for sizing. The two main principles are Kelly calculations for optimal betting and Bayesian reasoning with updated conditional probabilities. The basic process is: i) get base rate of % calls right (if you have a track record, use that, otherwise 50-50 or 40-60 would be fine). I also then have my own base rights for % gain when correct and % loss when incorrect ii) you can optimise with Kelly criteria around those base rates and you can back-calculate what position sizes you can maximally take. With my own base rates, taking up to 20% cost-base positions have a <1% probability of causing 30% portfolio drawdowns. I mainly focus on downside risk on larger positions- if you are familiar with the mathematics of Kelly's calculations, it is self-evident.
Often if you are going to 20% it is a fantastic idea and it promptly goes up a lot (every time I have gone to 20% it has been at least a 3x, but I think it has only been on 3-4 occasions that I have actually ever done it). For me, I don't actually care too much about the % after unrealised gains- I only care about the cost base. I often take some off the table at times for my own piece of mind for very large positions. This creates a new problem- long-lead time errors. If for instance you buy a "stonk" based on a bull-market momentum strategy, and then it rises 50 or 75%, and you log it as a correct call and use it to inform your base rates, you may get very false readouts. This is a very big problem. The only way you can be 100% sure that your unrealised gains were actually good calls, is when you have held through a market bottom and the bottom price was well above your cost base. I have now been through at least 2 bottoms, the one in late 2018 early 2019 and Covid, and in both cases my cost bases were ok. Ie I was not making structurally riskier bets at the top of the cycle.I am still quite conscious of this problem, but am somewhat reassured by my overall performance- I have been keeping detailed quarterly self-audit records for the last 14 quarters, and only underperformed the index in one quarter (Dec 2018).
As such, although having these large positions is not well advised and against most of the dogma of portfolio construction, I have my own method which has so far worked ok for me. I am sure I will make a big mistake at some stage- but as with all things, I will take it as an opportunity to learn when it does. To anyone else who happens to read this post- this is obviously my own random method only NOT advice, DYOR etc etc-
ADH Price at posting:
$3.21 Sentiment: Buy Disclosure: Held
(20min delay)