Actionable insights on equities, fixed-income, macros and personal finance Start 14-Days Free Trial
Actionable investing insights Get Free Trial

Poker and Trading


Amit Varma’s required-reading post on The Beautiful Game of Poker has parallels in the stock market as well. A game involving both skill and luck, poker has both been relegated to a pure gamble, a game of nerves, a great return for the professionally diligent. And why I see parallels in the stock market is that the concepts are nearly the same.

Like in poker, trading involves odds and probabilities. Like Amit says if the odds favour you, you go in, otherwise you stay out. He mentions a hand where the pot is Rs. 1,000 and an opponent has just gone all in with Rs. 800. Let’s say half of that pre-all-in pot was Amit’s – Amit stands to lose Rs. 500 on folding, but that shouldn’t be considered – since it’s history and he will lose it anyway if he bets and loses. If he has to win, he needs to put in Rs. 800 more, to win the Rs. 1,800. That’s odds of about 2:1 – and his probability of winning is 36%.

The expectancy of this offer is: (Amount you win)x(probabiilty of winning) – (Amount you lose)x(Probability of Losing). In this case it’s 1800 x 0.36 minus 800 x 0.64 or a positive figure, +136.

If the opponent had bet Rs. 2,000 more (instead of 800), then the Expectancy would be

3000×0.36 – 2000*0.64 = –200

A negative expectancy is not worth the effort – although once in a while Amit can go in, and get lucky (36% is a good enough chance). An amateur player would look at such lucky deals and call it skill – that he just “knew” it would happen and so on. But players like Amit would only go there if there was something else that changed the odds – for instance some idea that the opponent is bluffing, where suddenly the 36% chance of beating the opponent becomes 75%; again, he may be unlucky but he plays when the odds favour him.

(For the record, the optimum bet of the opponent in the earlier example is Rs. 1285. That straightens the odds entirely. I’m assuming poker players have most of that worked out, it’s a pure math concept.)

Now look at the game. A one off game has very little skill. You could do all the math you want but the odds are useless (unless they are like 99.99% in your favour). Play five games and you might get slightly better using the odds – but even there the odds were in your favour probably only once, and you got to play just once, which is not enough.

Play 500 games, and now we’re starting to see results. Losing hands against the odds becomes lesser and lesser because as the sample size grows the odds should work. (Note to real world: if the odds are not working after a long time, the game is rigged.)

Basically you have to find an edge – playing only when the expectancy is positive – and then play hajaar games. Will you still win?

Someone famous said, if you don’t bet, you can’t win. If you lose all your chips, you can’t bet. The idea then is to be able to play hajaar games without losing all your chips.

Let’s say your average game is Rs. 1,000 and in one hand, the probability of your winning is 80%, the pot has gone nuts, to say 100,000 and you need to bet Rs. 50,000 to win – but you have only Rs. 50,000 left. The expectancy is positive. But here’s the thing – if you bet and lose, you won’t have enough left to play the next hand. Would you still bet?

You can’t play if you bet and lose. If you bet and win, that’s the best result and you get to continue playing. But if you fold, on average you can play 50 more hands, and with your edge you’ll win some of them. I’d say fold and move on. But if a player goes on and wins, is he playing the odds, or was there more bravado? If he loses, should he have played the odds?

In poker this is tested often – you have to play the odds, they say. There’s two reasons for that. If an opponent knows your strategy is to continue to play, he’ll just keep putting in money until your filter triggers, and then grab the pot because you fold. So you play the odds; and you only use a small portion of your bankroll per game. So a guy playing with Rs. 50,000 will actually have 10 lakhs as a bankroll, just will not lose more than 50,000 in any one session. Then he can play such suicide odds.

In trading, this is exactly the thought process. If you find an edge – whether it’s value investing, or technicals, or algo trading, the idea is simply to find trades where you expectancy is positive and then make as many trades as possible. Take a simple strategy – buying all-time-highs, with a 20% stop loss. When you back test this strategy over a few markets, you might find that it yields a positive expectancy of say 5% per trade. Now the idea must be to trade all those markets – not just one. When stock markets are down, bonds are probably making highs. If not,commodities. Or a currency. Basically where you don’t get enough entries to maximize your edge, you work with more markets, more instruments and so on.

This is why a lot of algo trading – risk-based – is day traded. When you take a risk and do multiple trades every day, you can make your edge work for you in as little as a month. For a trader, this is a diamond – getting positive cash flow from your work in a month is fantastic. Some other traders care if they make a profit quarter on quarter – they might do about 10 trades a week then.

The edge – positive expectancy – is very difficult to quantify on an intraday basis. You don’t get fixed odds like poker. (The equivalent in poker is to have unlimited cards and the next card is completely random. You can visually try to guess but the odds are not quantifiable) But on a longer term as human psychology gets more embedded in the price, it’s easier to quantify using back-tests and so on. A number of statistical theories can be applied on a trading curve to see if it’s “stable” – that is, can it be expected to continue, given the same kind of wobbliness in input data going forward. Given we’re in the stock market, we have the time to analyze all such statistics and then take calls. (Poker has a faster turnaround time – imagine if I tried to open a charting application for each hand on a poker table)

Also the fundas are similar – if the odds work against you, you don’t trade. If you don’t find good odds, you don’t trade. If you need to take on a huge bet compared to your bankroll, you can choose not to bet (there’s no opponent bearing down on you). And better still, you can scale down position size with every trade. If I start with 1 lakh and put 5% on each trade (Rs. 5,000), and I lose say 6 trades continuously – remember, in small numbers it’s luck – I have only Rs. 70,000 remaining. I can then say I’ll still put only 5% and put in only Rs. 3,500 per trade. And so on. (In poker, stakes tend to be fixed per table and the jumps are sorta quantum, so you can’t do a granular amount of money management)

Lastly, over the long term, trading is a skill. The problem is that most people lose all their chips before they get skillful enough. And skill, as you might have realized, is as much about money management as about calculating the odds. The unfortunate part of the markets is that most people don’t even know the odds and have an even smaller idea of money management. And some of them still win. That’s because it takes just one big win. And that one big win can still be luck.


Like our content? Join Capitalmind Premium.

  • Equity, fixed income, macro and personal finance research
  • Model equity and fixed-income portfolios
  • Exclusive apps, tutorials, and member community
Subscribe Now Or start with a free-trial