**** The "Big Bet" Concept

[BigBollocks]

MOST, not all, but MOST professional handicappers I know flat bet virtually every bet they make (with the exception of maybe 1-5 games a year). It drives me nuts in terms of SBR pride (how little or much that is worth) to flat bet and go 6-1 at my sportsbook, put four of them up here and go 3-1, but only score one point on the week because the bonus shell I selected happened to lose.

Not only are you teaching bad money management to a lot of novice handicappers, you're making this somewhat of a crap shoot for a lot of the seasoned veterans. If I think I have an edge and like any bet enough to play it, it gets the same base unit play as any other play I make. Nothing like seeing a 6-6 handicapper who is down substantially at his/her sportsbook be ahead of a 9-3 handicapper who is up on the season due to this stupid "big bet" concept.

JJ's rant about the big bet concept and this contest in general might have been right on the money after all. I don't know what's more gay, this big bet concept or the Prick's fruity drink with an umbrella....

[jonmic]

If you don't like the concept of the contest, then don't play it. Simple.

[curious]

MOST, not all, but MOST professional handicappers I know flat bet virtually every bet they make (with the exception of maybe 1-5 games a year). It drives me nuts in terms of SBR pride (how little or much that is worth) to flat bet and go 6-1 at my sportsbook, put four of them up here and go 3-1, but only score one point on the week because the bonus shell I selected happened to lose.

Not only are you teaching bad money management to a lot of novice handicappers, you're making this somewhat of a crap shoot for a lot of the seasoned veterans. If I think I have an edge and like any bet enough to play it, it gets the same base unit play as any other play I make. Nothing like seeing a 6-6 handicapper who is down substantially at his/her sportsbook be ahead of a 9-3 handicapper who is up on the season due to this stupid "big bet" concept.

JJ's rant about the big bet concept and this contest in general might have been right on the money after all. I don't know what's more gay, this big bet concept or the Prick's fruity drink with an umbrella....

Flat betting is ridiculous. You bet your expected value. If a game has a higher probable outcome than normal you bet more on that game. I would never flat bet.

My college dog picks on Saturday weren't that impressive win loss wise, but they earned 77 units because I don't flat bet.

I know lots of professional gamblers and I don't know any that flat bet.

[slacker00]

I'm a flat bettor. I know the discussion will continue to rage between flat bettors & big bettors as long as there are bets to be taken. This topic emerges here on SBR at least every couple months.

The bottom line is the mathematics of money management. If you are risking 25% of your betting bankroll on a single play, there's a decent chance you'll go bankrupt if you make a play like that very often unless you can assess your chances on a 50/50 wager at 95% or better, which is simply unrealistic on widely available public lines. The 10-unit "lock" is a simple example of someone who is clueless about bankroll management.

That said, varying a one unit system to allow an occasional 2 unit play is an acceptable variation that is fine for a contest like BTP, IMHO. It's a healthy "compromise" between the two schools of thought. It allows for a big bet, while still maintaining a majority of single unit plays as well as avoiding the dreaded "10 unit lock".

[curious]

I'm a flat bettor. I know the discussion will continue to rage between flat bettors & big bettors as long as there are bets to be taken. This topic emerges here on SBR at least every couple months.

The bottom line is the mathematics of money management. If you are risking 25% of your betting bankroll on a single play, there's a decent chance you'll go bankrupt if you make a play like that very often unless you can assess your chances on a 50/50 wager at 95% or better, which is simply unrealistic on widely available public lines. The 10-unit "lock" is a simple example of someone who is clueless about bankroll management.

That said, varying a one unit system to allow an occasional 2 unit play is an acceptable variation that is fine for a contest like BTP, IMHO. It's a healthy "compromise" between the two schools of thought. It allows for a big bet, while still maintaining a majority of single unit plays as well as avoiding the dreaded "10 unit lock".

Interesting, you use the word "lock", which almost everyone here disdains, to taint the idea of varying bet size to be some factor of expected value times bankroll. I would agree that randomly betting a large number of units without knowing what the true expected value of the bet is totally idiotic. I don't use the word "lock" and no one I know uses that word.

You are confusing making bet size equal a percentage of bankroll based on expected value with overbetting. The first step in knowing the bet size is using something like kelly criterion (kelly isn't the only way to do this) to determine bet size based on probability of outcome and odds being offered by the book. The second step is to reduce the bet size given by kelly (or whatever other formula you use) to keep the risk of ruin at an acceptable limit. Kelly ignores variance, variance will either give you a nervous breakdown or devestate your bank. Personally, I use a risk of ruin of 1%.

I would NEVER bet 25% of my bankroll on a single bet. I would never bet anywhere close to 25% of my bankroll on a single bet.

If you are a flat bet holy warrior then you obviously know nothing about optimizing bankroll growth, and your protestations to the contrary, you don't follow good money management practices because money management is ONLY about optimizing bankroll growth while at the same time keeping the risk of ruin very low. Anyone who does that would never bet a high % of their bankroll on a single bet.

[Doc JS]

MOST, not all, but MOST professional handicappers I know flat bet virtually every bet they make (with the exception of maybe 1-5 games a year). It drives me nuts in terms of SBR pride (how little or much that is worth) to flat bet and go 6-1 at my sportsbook, put four of them up here and go 3-1, but only score one point on the week because the bonus shell I selected happened to lose.

Not only are you teaching bad money management to a lot of novice handicappers, you're making this somewhat of a crap shoot for a lot of the seasoned veterans. If I think I have an edge and like any bet enough to play it, it gets the same base unit play as any other play I make. Nothing like seeing a 6-6 handicapper who is down substantially at his/her sportsbook be ahead of a 9-3 handicapper who is up on the season due to this stupid "big bet" concept.

JJ's rant about the big bet concept and this contest in general might have been right on the money after all. I don't know what's more gay, this big bet concept or the Prick's fruity drink with an umbrella....

As has been previously stated, the flat bet vs. big bet debate goes on here just about as much as What's your favorite sports book.

I think most, if not all of us here, realize that there is a fair amount of luck involved with the BTP contest. Yesterday, Mizzou scored a "meaningless" TD in the final seconds to cover for me by 1/2 point and make me a winner in BTP for the second week in a row.

I do not think that SBR is "advocating" the big bet concept. It's just part of the contest. It's free. If you don't like it - don't play. Simple as that.

[moneyplays]

I love the big bet concept. It makes you decide on which game you feel most confident with.

[BigBollocks]

Great discussion on both sides guys. Curious I'd like to fancy that I have a fairly strong statistics background, but the determination on how strong of an advantage you have in any one particular event is sheerly subjective. Slacker is correct that for 95% of the handicapping population varying bet sizes is a dangerous proposition that cannot be managed. I have a fairly rigorous standard on all selections that I make, and flat bet and re-adjust from there.

I find it hard to believe that you know "a lot" of professionals and don't know one that doesn't flat bet, but to each their own. You're obviously an intelligent guy (and constantly make it apparent to everyone else just how smart you believe yourself to be), so I have no doubt you're capable of coming out ahead using a variety of techniques.

This debate has raged between handicapping number guys for years, and will no doubt never fade away. GL...

[BigBollocks]

Oh, and since this is a contest forum, in terms of the contest can we at least have a vote next year on whether or not to keep the big bet concept intact? Cheers...

[curious]

Great discussion on both sides guys. Curious I'd like to fancy that I have a fairly strong statistics background, but the determination on how strong of an advantage you have in any one particular event is sheerly subjective. Slacker is correct that for 95% of the handicapping population varying bet sizes is a dangerous proposition that cannot be managed. I have a fairly rigorous standard on all selections that I make, and flat bet and re-adjust from there.

I find it hard to believe that you know "a lot" of professionals and don't know one that doesn't flat bet, but to each their own. You're obviously an intelligent guy (and constantly make it apparent to everyone else just how smart you believe yourself to be), so I have no doubt you're capable of coming out ahead using a variety of techniques.

This debate has raged between handicapping number guys for years, and will no doubt never fade away. GL...

How could you possibly know who I know or who I don't know?
I don't understand your slight. I post picks and the unit sizes and I explain why i vary the unit sizes the way I do. Anyone who has followed my picks in both baseball and college football is up tons. I know a lot of people who make a good living at some form of gambling whether it be blackjack, sportsbetting, or horse racing, none of them flat bet. It isn't that difficult to prove that betting true expected value is far superior in terms of bankroll growth to flat betting. In one week my picks earned 90 units profit. Regarding your "...will be debated forever...", I don't like philosophical debate, it is a waste of time. If you can show that flat betting is a better means of optimizing bankroll growth than betting true expected value, then I would use it. I would agree that if a bettor does NOT know what the true expected value is then they should flat bet.

[Ganchrow]

It isn't that difficult to prove that betting true expected value is far superior in terms of bankroll growth to flat betting.

I would say you'd have a very difficult time proving this assertion as, in general, it is false. It may frequently be true, but it is not always true.

What is true in general is that betting the Kelly stake will always be superior in terms of bankroll growth to flat betting, and would also always be superior to betting a fraction of bankroll equal to the EV (except in the case of even odds, when the two would be equivalent).

In order to maximize bankroll growth a bettor needs to bet to win a percentage of his bankroll equal to his EV. If a bettor risks a percentage of his bankroll equal to his EV, he'll be betting too much on underdogs and not enough on favorites. So in other words, with an edge of 5%, one would bet 10% of bankroll at -200, or 2.5% of bankroll at +200.

[curious]

I would say you'd have a very difficult time proving this assertion as, in general, it is false. It may frequently be true, but it is not always true.

What is true in general is that betting the Kelly stake will always be superior in terms of bankroll growth to flat betting, and would also always be superior to betting a fraction of bankroll equal to the EV (except in the case of even odds, when the two would be equivalent).

In order to maximize bankroll growth a bettor needs to bet to win a percentage of his bankroll equal to his EV. If a bettor risks a percentage of his bankroll equal to his EV, he'll be betting too much on underdogs and not enough on favorites. So in other words, with an edge of 5%, one would bet 10% of bankroll at -200, or 2.5% of bankroll at +200.

As always you leave out variance. Using full Kelly in theory is great. But computers don't suffer from nervous breakdowns when Kelly lets their bankroll go from $10,000 to .01 just before it goes back to $20,000. Variance is what knocks most bettors out of the game.

I disagree with your assertion that risking EV is betting too much on underdogs. I don't know how you are defining EV but I define it as true probability of outcome x odds. You are suggesting that the underdogs have a lower probability of outcome, which is true in some cases, not in all cases, but the actual probability has already been factored in.

I like the way you use nomenclature to make yourself look smart when you know from having read my earlier posts that you are not talking about what I said earlier because I already defined what I meant by true expectation. Probably I am not using the "text book" definition, but I defined the formula I used earlier for true expectation, and I know you understand it. So you are saying the same thing I said except for the fact that you don't tell people the risks of using full Kelly, but you are using big words to make it look like you are smarter. Did you not get love as a child?

[swifty]

I love the big bet concept. It makes you decide on which game you feel most confident with.

I agree with you, You can bet or not bet, but for oe I know to bet low on about 5 - 10 teams you could be 6-4 or not. but its the fun of watching your team play on TV and betting them.

[RickySteve]

As always you leave out variance. Using full Kelly in theory is great. But computers don't suffer from nervous breakdowns when Kelly lets their bankroll go from $10,000 to .01 just before it goes back to $20,000. Variance is what knocks most bettors out of the game.

I disagree with your assertion that risking EV is betting too much on underdogs. I don't know how you are defining EV but I define it as true probability of outcome x odds. You are suggesting that the underdogs have a lower probability of outcome, which is true in some cases, not in all cases, but the actual probability has already been factored in.

I like the way you use nomenclature to make yourself look smart when you know from having read my earlier posts that you are not talking about what I said earlier because I already defined what I meant by true expectation. Probably I am not using the "text book" definition, but I defined the formula I used earlier for true expectation, and I know you understand it. So you are saying the same thing I said except for the fact that you don't tell people the risks of using full Kelly, but you are using big words to make it look like you are smarter. Did you not get love as a child?

You're not much of a Winston Churchill fan, are you, curious?

[Ganchrow]

As always you leave out variance. Using full Kelly in theory is great. But computers don't suffer from nervous breakdowns when Kelly lets their bankroll go from $10,000 to .01 just before it goes back to $20,000. Variance is what knocks most bettors out of the game.

No. I don't leave out variance. Nor does Kelly. I specifically addressed this fallacious notion in my last post to you on the topic. If you care to address the substance of what I've written without simply repeating yourself or resorting to ad hominems, I'd be more than happy to continue our conversation.

I disagree with your assertion that risking EV is betting too much on underdogs. I don't know how you are defining EV but I define it as true probability of outcome x odds.

With all due respect, your comments belie any understanding you claim to have of the conclusions of Kelly. Disagree with my assertion if you so desire, but realize it's not only my assertion, but Kelly's as well. (If you care to read the original Kelly paper, I've archived a PDF here.) If you care to read a simple mathematical proof of said assertion, check out the 2nd footnote of this article I wrote on Kelly, Maximizing Expected Growth (Kelly criterion Part II).

Now if you have a mathematical reason for disagreeing with this, devoid of ad hominems, I'd be pleased to hear it and would respectfully respond. (By the way you might also want to check out that article in its entirety. I do address some of the points you've made.)

You are not talking about what I said earlier because I already defined what I meant by true expectation. Probably I am not using the "text book" definition, but I defined the formula I used earlier for true expectation, and I know you understand it.

Your definition of expectation is indeed correct. That said, when you state that an advantage player should wager his expectation times his bankroll, I think you mean to say that he should wager his expectation minus one times his bankroll, as expectation will always be greater than unity for any with positive edge. Nevertheless, I do understand your point, and that point is incorrect. The full Kelly formula is:

(prob*odds - 1)/(odds-1)

Where prob corresponds to what you call the "true probability", and odds to the decimal odds. Hence, the numerator is the EV (or the EV - 1 ), depending upon how you look at it, and the denominator corresponds to the net [ayout one would receive on a single unit bet (i.e., 1 for odds of +100, 2 for odds of +200, 0.5 for odds of -200). Now if you believe this formula to be incorrect, I'd certainly be interested in hearing why. Realize that this formula is the same one derived in Kelly's original paper, is the same one I use in my Kelly calculator, is the same one referenced in the Wikipedia entry on the Kelly criterion, is the same one referenced in Beat the Dealer author Edward Thorpe's famous paper on Kelly, and is the same one used in the Wizard of Odds section on Kelly. Now I'll certainly admit the possibility that all of us are wrong and that you and you alone are right ... but come on. Why not just entertain what I've said? Just a bit. I promise you ... I do know a little about Kelly.

[dudefrompa]

......what the hell

Now that's complaining. Seriously, if you don't like the contest that much, then here's a thought.......DONT PLAY. All this talk about flat bets and bad for the betting community....you lost me. This isn't a sportsbook, it's a damn contest.