![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Bet sizing, compensation, accounting
Apr. 5th, 2009 01:11 pmOne of the strangest tendencies that I find in No Limit poker players is their attempt to find a bet size that will get worse hands to fold and better hands to call, and then to bet precisely that amount. Indeed, they are often proud of betting such an amount, so that they can "find out where they are".
I think that my increased profit over the past six months can be attributed to two main changes:
a) I have learnt to lay down hands on the river that before I would have called with.
b) I am winning more uncontested hands on the turn than I did before, and fewer on the flop.
And the reason for (b) is simple. I am raising slightly less pre-flop and betting slightly less on the flop. I follow this with a significantly bigger bet (as a proportion of the pot) on the turn. Winning hands on flop compared to winning pre-flop - more moneys. Winning hand on the turn = more moneys still. Indeed, I am thinking of going back to $50 BI 6-max to see whether I can take this principle further -- bet a slightly smaller amount (as a percentage of the pot) on the turn, and then throw in an overbet on the river. This will require some experimentation and is sure to go tits up on some boards and against certain types of player. But the rewards, if it can be got right, would be considerable.
These smaller raises pre-flop (basically, raise just enough to make people think it's a genuine raise rather than an attempt to see the flop cheaply) and small bets on the flop (once again, just enough to look like a genuine bet rather than one looking to see the turn cheaply) encourage weaker hands to see the flop or to peel one off.
Ah, the critics mnight say, but that means you don't know where you are and are often out of position.
And here we have to look at the stack sizes.
But also look at it this way. Would you rather your opponent only call your bet if he was beating you (certainty of loss) or that sometimes he call you with a worse hand and sometimes with a better one?
Big overbets are often similarly criticized because worse hands will always fold, but if you get better hands to fold a sufficient proportion of the time (where "sufficient" depends on a number of mathematical factors) then the big overbet is fine.
My current style is to mix and match the underbet and the overbet, throwing in the "right-size" bet only a proportion (25%ish, perhaps) of the time. This works at $100 and $200 BI's because most people are either serial underbettors or (more often) serial overbettors. And if they do mix them, it tends to be an overbet on the flop and an underbet on the turn, which is precisely the wrong way round.
+++++++++++++++++
The Treasury Select Committee gave sterling support (in the sense that sterling support is "strong" -- perhaps not a guarantee any more) to the UK government in one important sector in its report released on Friday: Point 88 (http://www.publications.parliament.uk/pa/cm200809/cmselect/cmtreasy/402/40208.htm#a32)
states that:
All very non-controversial, you might think. But, hold! As my mother observed a year or so ago, every time she went into the Post Office a couple of years ago, she was offered a "Post Office" savings scheme. What, she might think, could be more British than the Royal Mail?
Except that the Post Office isn't the Royal Mail, and the savings scheme that she was being offered was run by the Bank of Ireland. As the Chamber of Commerce stated, "The Post Office offer savings through Bank of Ireland where, in a financial meltdown you are likely going to have to claim from the Irish authorities."
This is a country that just had its credit rating downgraded to less than sparkling.
Let's have a look at a different aspect. Suppose you spread your risk around. You put some money in Abbey, some in Alliance & Leicester.
No good. Both are owned by Spain's Santander. Santander is registered with the FSA but, understandably, only once. Your Abbey and A&L accounts are counted as one under UK compensation rules.
Much of the fault here lies in the ability of companies to have a "brand" and an official company name. Aviva, for example, traded as Norwich Union here, Hibernian in Ireland, Commercial Union in Poland and Aviva elsewhere. And another fault is that companies aren't keen to tell you whether they are responsible for an account, or someone else is. My mother got hit by a car coming out of a side road a couple of weeks ago. The guy said that he was insured with the AA. Now, don't criticize him. Where do you see in the AA adverts any loud phrase along the lines of "actually, we just pass the insurance onto someone else"? But that's what they do. In this case, his insurer was Zurich.
If the government and the TSC are going to say that it's an important principle that the UK not guarantee foreign deposits, thhey should also ensure that punters get a reaonable crack of the whip (outside the small print) in finding out wherre their money is actually going.
++++++++++++
Some time ago I wrote that one neat way out of the current capitalization crisis would be to relax the rules on capital. At the moment we are looking at something vaguely impersonating the pushmepullyu in Dr Doolittle. (WARNING. THIS MIGHT GET TECHNICAL...)
On Thursday in the US there was a change to accounting rules. A few years ago the famous "mark-to-market" valuation rules were introduced in the US. They were tweaked to be more strict in late 2006 or thereabouts. The major problem with this was that it was assumed that there would a market which you could mark to. For much of the time in the past 18 months, several products have effectively been unsaleable. This forced huge valuation write-offs. So, the banks went to Congress and complained. Congress caved in and threatened the Financial Accounting Standards Board, which sets the accounting rule.
The FASB caved in on Thursday, with just a minor "transparency" sop to investors.
In effect, banks can (indeed, must) now value assets that they have no intention of selling at far closer to estimated maturity value, provided the assets are "performing" (i.e., interest is being paid).
So, five accounting guys in Connecticut have basically turned the accounting world (well, the US accounting world) upside down, eliminating the need for write-offs on loan books, products, whatever, so long as (a) the product is a strategic holding and (b) the product hasn't defaulted.
This saves the arse of many many companies and is, in effect, a relaxation of capitalization rules.
And yet, the day before, Sheila Bair, chairman of the Federal Deposit Insurance Corp, was a system that would RAISE the capital requirements of banks. Then, on the same day that the FASB was reducing the pressure on banks, the Financial Stability Forum accepted for the first time the idea of a capital floor for banks, referring to it as a "supplementary non-risk based measure to contain bank leverage."
Then it occurred to me that this apparent contradiction is actually nothing of the kind. If you change the accounting rules, then you can make the capital requirements LOOK tougher, when in fact they are more relaxed.
How so?
Well, suppose you have a loan out. The debtor is servicing that loan, but because of the credit crisis, such loans are selling in the market at a steep discount. This might be because there are forced sellers of other, similar such loans. However, you are not a forced seller. But, under the most recent accounting rules, your loan "asset" has to be marked at a value that it is being traded at. Indeed, another bank could stitch you up by "trading" such loans at an artificially low price.
Since its value has fallen, your capital has fallen, and so you need to strengthen your capital.
Now, suppose you have already done this (you might have done it last October, amidst the Lehman Bros fallout). Today, the regulators demand that you hold higher capital. Simultaneously, the accounting rules are relaxed that boost your capital by far more than this.
What looks like a toughening of the regime is in fact a relaxation. The FDIC demands $2bnn of capital rather than $1bn, while the FASB revalues your $1.2bn of capital to $2.4bn. What had been a $200m surplus becomes a $400m surplus, giving you an extra $2000m which you can now lend out to new customers.
Clever, or what?
________________
I think that my increased profit over the past six months can be attributed to two main changes:
a) I have learnt to lay down hands on the river that before I would have called with.
b) I am winning more uncontested hands on the turn than I did before, and fewer on the flop.
And the reason for (b) is simple. I am raising slightly less pre-flop and betting slightly less on the flop. I follow this with a significantly bigger bet (as a proportion of the pot) on the turn. Winning hands on flop compared to winning pre-flop - more moneys. Winning hand on the turn = more moneys still. Indeed, I am thinking of going back to $50 BI 6-max to see whether I can take this principle further -- bet a slightly smaller amount (as a percentage of the pot) on the turn, and then throw in an overbet on the river. This will require some experimentation and is sure to go tits up on some boards and against certain types of player. But the rewards, if it can be got right, would be considerable.
These smaller raises pre-flop (basically, raise just enough to make people think it's a genuine raise rather than an attempt to see the flop cheaply) and small bets on the flop (once again, just enough to look like a genuine bet rather than one looking to see the turn cheaply) encourage weaker hands to see the flop or to peel one off.
Ah, the critics mnight say, but that means you don't know where you are and are often out of position.
And here we have to look at the stack sizes.
But also look at it this way. Would you rather your opponent only call your bet if he was beating you (certainty of loss) or that sometimes he call you with a worse hand and sometimes with a better one?
Big overbets are often similarly criticized because worse hands will always fold, but if you get better hands to fold a sufficient proportion of the time (where "sufficient" depends on a number of mathematical factors) then the big overbet is fine.
My current style is to mix and match the underbet and the overbet, throwing in the "right-size" bet only a proportion (25%ish, perhaps) of the time. This works at $100 and $200 BI's because most people are either serial underbettors or (more often) serial overbettors. And if they do mix them, it tends to be an overbet on the flop and an underbet on the turn, which is precisely the wrong way round.
+++++++++++++++++
The Treasury Select Committee gave sterling support (in the sense that sterling support is "strong" -- perhaps not a guarantee any more) to the UK government in one important sector in its report released on Friday: Point 88 (http://www.publications.parliament.uk/pa/cm200809/cmselect/cmtreasy/402/40208.htm#a32)
states that:
We agree that the overarching principle should be that the UK Government cannot provide cover for deposits held by British citizens in jurisdictions outside the direct control of the United Kingdom.
All very non-controversial, you might think. But, hold! As my mother observed a year or so ago, every time she went into the Post Office a couple of years ago, she was offered a "Post Office" savings scheme. What, she might think, could be more British than the Royal Mail?
Except that the Post Office isn't the Royal Mail, and the savings scheme that she was being offered was run by the Bank of Ireland. As the Chamber of Commerce stated, "The Post Office offer savings through Bank of Ireland where, in a financial meltdown you are likely going to have to claim from the Irish authorities."
This is a country that just had its credit rating downgraded to less than sparkling.
Let's have a look at a different aspect. Suppose you spread your risk around. You put some money in Abbey, some in Alliance & Leicester.
No good. Both are owned by Spain's Santander. Santander is registered with the FSA but, understandably, only once. Your Abbey and A&L accounts are counted as one under UK compensation rules.
Much of the fault here lies in the ability of companies to have a "brand" and an official company name. Aviva, for example, traded as Norwich Union here, Hibernian in Ireland, Commercial Union in Poland and Aviva elsewhere. And another fault is that companies aren't keen to tell you whether they are responsible for an account, or someone else is. My mother got hit by a car coming out of a side road a couple of weeks ago. The guy said that he was insured with the AA. Now, don't criticize him. Where do you see in the AA adverts any loud phrase along the lines of "actually, we just pass the insurance onto someone else"? But that's what they do. In this case, his insurer was Zurich.
If the government and the TSC are going to say that it's an important principle that the UK not guarantee foreign deposits, thhey should also ensure that punters get a reaonable crack of the whip (outside the small print) in finding out wherre their money is actually going.
++++++++++++
Some time ago I wrote that one neat way out of the current capitalization crisis would be to relax the rules on capital. At the moment we are looking at something vaguely impersonating the pushmepullyu in Dr Doolittle. (WARNING. THIS MIGHT GET TECHNICAL...)
On Thursday in the US there was a change to accounting rules. A few years ago the famous "mark-to-market" valuation rules were introduced in the US. They were tweaked to be more strict in late 2006 or thereabouts. The major problem with this was that it was assumed that there would a market which you could mark to. For much of the time in the past 18 months, several products have effectively been unsaleable. This forced huge valuation write-offs. So, the banks went to Congress and complained. Congress caved in and threatened the Financial Accounting Standards Board, which sets the accounting rule.
The FASB caved in on Thursday, with just a minor "transparency" sop to investors.
In effect, banks can (indeed, must) now value assets that they have no intention of selling at far closer to estimated maturity value, provided the assets are "performing" (i.e., interest is being paid).
So, five accounting guys in Connecticut have basically turned the accounting world (well, the US accounting world) upside down, eliminating the need for write-offs on loan books, products, whatever, so long as (a) the product is a strategic holding and (b) the product hasn't defaulted.
This saves the arse of many many companies and is, in effect, a relaxation of capitalization rules.
And yet, the day before, Sheila Bair, chairman of the Federal Deposit Insurance Corp, was a system that would RAISE the capital requirements of banks. Then, on the same day that the FASB was reducing the pressure on banks, the Financial Stability Forum accepted for the first time the idea of a capital floor for banks, referring to it as a "supplementary non-risk based measure to contain bank leverage."
Then it occurred to me that this apparent contradiction is actually nothing of the kind. If you change the accounting rules, then you can make the capital requirements LOOK tougher, when in fact they are more relaxed.
How so?
Well, suppose you have a loan out. The debtor is servicing that loan, but because of the credit crisis, such loans are selling in the market at a steep discount. This might be because there are forced sellers of other, similar such loans. However, you are not a forced seller. But, under the most recent accounting rules, your loan "asset" has to be marked at a value that it is being traded at. Indeed, another bank could stitch you up by "trading" such loans at an artificially low price.
Since its value has fallen, your capital has fallen, and so you need to strengthen your capital.
Now, suppose you have already done this (you might have done it last October, amidst the Lehman Bros fallout). Today, the regulators demand that you hold higher capital. Simultaneously, the accounting rules are relaxed that boost your capital by far more than this.
What looks like a toughening of the regime is in fact a relaxation. The FDIC demands $2bnn of capital rather than $1bn, while the FASB revalues your $1.2bn of capital to $2.4bn. What had been a $200m surplus becomes a $400m surplus, giving you an extra $2000m which you can now lend out to new customers.
Clever, or what?
________________
