Double It Up

[peterbirks]

An interesting entry from ex-Pokerstars employee Terrence Chan (terrencechan.livejournal.com/) on a roller-coaster ride against a strange opponent.

Basically this guy, who is apparently a poor heads-up player, sits down with $1,000, and promptly loses it. Nothing unusual there. But what is interesting is that, if he wins, he wants to move up to a higher level. Terrence has been happily accommodating this request, eventually busting him every time, 19 times on the spin.

Except that the last time, the guy's run of fortune (which can last some time at heads-up) kept continuing. And he kept asking to double up, until he got to saying that he wanted to play $500-$1,000. Terrence, stuck for some $21,000, was understandably hesitant to risk playing at stakes he was not comfortable with.

But, and here's the beautiful catch, as soon as Terrence refused to move up, the guy left!

What a dilemma, huh? "I'll give you a chance to win your money back, but only at higher stakes"
"But those stakes are too high!"
"Tough, who said life was fair?"

It all had a happy ending. Terrence swallowed hard, moved up in stakes, and busted the guy, by about 6am.

But you have to admire the maniac. He gets a lot of value for his $1,000 a week, and he can probably afford it. And, one day, he'll double-up (through sheer good fortune) in a way that will bust even the best heads-up player in the world. That, or the heads-up opponent will have to swallow his pride and accept his loss thus far. But players willing to move up to those high stakes wouldprobably prefer to risk being totally busted than to refuse the chance to play at higher stakes at positive EV.

A Fascinating Tale

Comments

[jellymillion.livejournal.com]

...players willing to move up to those high stakes would probably prefer to risk being totally busted than to refuse the chance to play at higher stakes at positive EV.

This was one of the things that Beal was (and is again) testing when he made the Big Game look like penny-ante schoolboy poker, was it not? Take the game to a level where the "better" player (not sure why I quoted that, Terrence is scarily good) is suddenly having to risk serious financial pain and knows they're at the mercy of short-term variance because they don't have the bankroll at those stakes to absorb a reverse.

So the pro is removed from the long-term +EV comfort zone and pushed into utility function territory, to the possible detriment of their game, which feeds back into EV and a potentially nasty loop develops.

One of the posts of the month.

[peterbirks.livejournal.com]

Well, there are obviously elements of Bealism here -- pushing the opponent out of his comfort zone.

I'm not reading new poker books on no-limit tournaments, but pushing your opponents out of their comfort zone is an important strategy and I hope that someone is writing about it.

The difference is that (a) Beal started with a lot of money, whereas this guy seems to be punting "just" $1,000 a week or thereabouts, (b) Beal accepted it when his opponents refused to move up in stakes, and (c) by playing the "double-the-stakes or I leave" ploy, the maniac is effectively "forcing" the opponent to play the Martingale, albeit one where the odds are on your side rather than against you.

The interesting thing for me is the refusal by Terrence to walk away from positive EV, even though it looks as if, from a bankroll management point of view, he should have walked away at, I dunno, I guess at a $10,000 loss? That would still leave Terrence $9,000 up overall and perhaps the next time, the guy would be back to $1,000 and playing at the original stakes.

Now, we don't know how much of Terrence's total bankroll was at risk. Does he own his flat outright? How much of his lifetime profit has left the poker economy never to return (now sitting in stocks, investment plans, etc)? We don't know.

Notwithstanding all that, if Terrence were to get busted, he could grind it back at 30-60 over three to six months. No fun, but not terminal. In other words, getting busted, as with losing an important hand in a tournament, is nowhere near as terminal as people might think. Getting busted in real-life cash rather than poncy tournaments is more important, but you aren't out of the game forever.

But, well, I would have walked away when 5K down. But, and this was my point, this is the difference. People with the guts and the ability to get to stratospheric stakes are not the type to say, "oooh, I'm 5K down. I know that I'm better than this guy, but maybe it isn't my night".

Like hell. They are saying "I am better than this guy. I am going to bust hime, even if it takes all my bankroll and all night to do it".

And, well, it did.

That reminds me...

[(Anonymous)]

... of my late father's roulette system, which I think he explained to my mother sometime in the early 1960s.

The system was quite simple. Make a low-risk bet (black or white?); if you lose, repeat the bet with double the stake. Go on doubling until you win.

This 'system' doesn't and can't change the mathematics of roulette: every time you bet, your expectation is negative. But the system will give you a lot of small wins, occasionally punctuated by an enormous loss when you can't afford to double up any further. The enormous loss will of course wipe out all your previous winnings, so I don't see much point in it.

If you're really unlucky, you might get the enormous loss the first time you try it...

I don't think my father ever tried it. He was living on overdraft most of the time, and fortunately he had enough sense not to go out gambling with borrowed money.

-- Jonathan

Re: That reminds me...

[(Anonymous)]

Your father hadn't found a new method. What he was suggesting is the martingale system that Pete refers to in an earlier comment. Hundreds of variations on the same theme exist, but all of them fall foul of the basic problem that you correctly point out.

Re: That reminds me...

[peterbirks.livejournal.com]

Yes, I called it the Martingale, because I assumed that most readers of this blog would know of the "system".

There is a "good" system for roulette (well, a "least bad" system). What you do is work out how much you would have placed on a roulette wheel throughout your entire life.

Save that money up. Then walk into a casino and place it all on Red.

This gives you the best rate of return against the casino at the roulette table. It does not, however, give you the best chance of a win. For the best chance of a win, you need to divide this money up in such a fashion that it can be bet Martingale style. This shifts you towards lots of small wins, but one very big loss.

The maniac, as mentioned in our first post, has found a "spin-up" system that seems to guarantee him lots of small losses, but the (negligible, but real) chance of a very big win -- plus, of course, the kudos of saying either that "X is too scared to play me; he quit" or (less likely ) "I busted X!"

Re: That reminds me...

[(Anonymous)]

Yes, I didn't recognize the Martingale name, but I never imagined that my father invented it. I just meant that this kind of thing has been around for a long time.

-- Jonathan

You mean we need to pass an exam to read this stuff?

[(Anonymous)]

"I assumed that most readers of this blog would know" -- hm, when you put something on the Web you can get readers of all kinds, including the man on the Barcelona omnibus.

I'm happy to admit that I know almost nothing about poker and very little about gambling in general. I came here in the first place because you invited me and because we've known each other for a long time. But I read this blog despite the heavy poker content, not because of it.

Usually I pass over the poker in silence, but this particular piece rang a bell with me, so I mentioned the fact.

-- Jonathan

Re: You mean we need to pass an exam to read this stuff?

[peterbirks.livejournal.com]

Oh, I'm not saying that I was right to assume that most readers would know what the Martingale theory was, just that I did so.

There is always a problem with writing in that you have to assume certain things on the part of the readership, otherwise you have the equivalent of "reinventing the wheel" every time you write anything (and, remember, these things tend to be written very much at speed).

You might be interested to note that I have, in several places, put in explanations that your bog-standard poker blog most definitely would not put in, because I know that there are readers from outside the poker blogosphere.

But the Martingale, well, it's mathematical, although I accept that its best-known application is to gambling. It was for this reason that I probably skipped over it. (Just googled it, and the top reference is to "mathworld", although my case is somewhat weakened by the fact that subsequent references are from esteemed places such as casinogamblingworld...)

PJ

Re: You mean we need to pass an exam to read this stuff?

[(Anonymous)]

As usual, there's an article about it in Wikipedia: http://en.wikipedia.org/wiki/Martingale.

Indeed it is mathematical, although it seems that the concept and the name originated in the gambling world several centuries ago.

-- Jonathan

Re: That reminds me...

[(Anonymous)]

There is another system to beat roulette and this one is guaranteed to work. It involves an infinite sum of money, and a house that is willing to take any sum gambled, and after that the mechanics are quite easy. It's never actually been tried for real, which seems a shame, and I'm collecting towards realising the dream. Spare a tenner?

[terrencechan.livejournal.com]

fwiw it's definitely more than $1000/week. He plays most days, I think, and $1000 is just what he'll risk per session. If (when) he loses the initial $1k stake, he'll virtually always reload for another dime. Again and again.