How to deal with this staking problem?

[Gaze73]

Step 1: Run hot and double your bankroll
Step 2: Bet 50-100% higher stakes
Step 3: Lose all profits on a cold run
Step 4: Back to square one.

The only solution seems to be to always bet the same no matter how much your BR grows but that only works with huge BRs. It's frustrating because even with a huge edge you can't win when the losses are 50-100% bigger than wins.

[Vyasports]

I know that feel bro...

[Gaze73]

Flat stakes BR: 1k to 2k to 1.5k to 2.5k to 2k
Doubled stakes BR: 1k to 2k to 1k to 2k to 1k

On one hand every hot run is followed by a cold run, so you shouldn't double your stakes. On the other hand gambler's fallacy says that cold and hot runs are completely unrelated so you should double your stakes. Damned if you do, damned if you don't.

[Shutup]

It is rough because people tend to bet more as their BR improves

Lets assume you start with 1,000
Bet 100 a game
You go 20-10 for a profit of 900
BR is now 1900 causing you to bet 200 a game
You then go 12-18 causing you to lose 1460

You went 32-28 which should give you profit but instead you are down 560

[Gaze73]

Kelly says you should increase your stakes.

Also the problem works in reverse. You start with a cold run and go from 1k to 500. If you halve your stakes, as you should according to Kelly, a hot run with lower stakes will make you just break even instead of making a nice profit.

Bookies with their infinite bankrolls have a huge advantage even if you can beat the juice. It's like a poker game, even the best player in the world playing against an average joe will have trouble winning with 1k chips against 1 million.

[KVB]

I think you don't have a staking problem, I thnk you could be creating one though.

Changing your bets size will increase vigorish.

If you are making some bets that are twice other bets, only the bigger bets are going to matter.

You could try a partial Kelly, but if think "running hot" indicates an edge then you could be sadly mistaken. Any form of Kelly will decimate you.

Keep your risk flat, and a small percentage of bankroll (like 1 or 2) and when you've increased your bankroll roughly 50%, or decreased roughly 25%, then readjust your 1 or 2%.

I think this style would be best for a bettor of your level (I am assessing your level based on the posts in this thread).

Good Luck Gaze, I know you have more experience than you let on and I know that you know trying to increase your bet size to somehow increase your advantage is dangerous at the least, and not really possible at best.

[Gaze73]

I definitely have an edge because I track all my picks in spreadsheets. Paper profits on flat stakes are constantly going up month after month but the real bankroll is stagnating because I always catch an absolute nightmare run every time I up the stakes. Every hot run gives me confidence that the edge is sustainable long term so there is no reason not to raise the stakes, and yet a cold run comes every time.

[KVB]

I definitely have an edge because I track all my picks in spreadsheets. Paper profits on flat stakes are constantly going up month after month but the real bankroll is stagnating because I always catch an absolute nightmare run every time I up the stakes. Every hot run gives me confidence that the edge is sustainable long term so there is no reason not to raise the stakes, and yet a cold run comes every time.

Again, though, you may be creating the staking problem. It doesn't matter what edge you have, or how much you track in any spreadsheet, if you don't have the discipline to control your bets, especially the sizes, then you will fail. Period.

Why do you bet more after a run of wins? Why wouldn't you bet more after a run of losses?

See what I mean? You are gambling when you may not need to.

[Roger T. Bannon]

I definitely have an edge because I track all my picks in spreadsheets. Paper profits on flat stakes are constantly going up month after month but the real bankroll is stagnating because I always catch an absolute nightmare run every time I up the stakes. Every hot run gives me confidence that the edge is sustainable long term so there is no reason not to raise the stakes, and yet a cold run comes every time.

You do not know your edge. If you do not know your edge, you will not win.

[Gaze73]

Again, though, you may be creating the staking problem. It doesn't matter what edge you have, or how much you track in any spreadsheet, if you don't have the discipline to control your bets, especially the sizes, then you will fail. Period.

Why do you bet more after a run of wins? Why wouldn't you bet more after a run of losses?

See what I mean? You are gambling when you may not need to.

It's not about discipline but about what's mathematically correct. Let's say that according to half kelly I should always bet 1%. After two weeks my BR doubles from $1k to $2k, and 1% is now 2% of the original BR. So my next bet according to half kelly should be $20, there is no reason not to bet $20. If I deposited $2k right away the bet would also be $20.
Betting more after a run of losses is gambler's fallacy. Just because you go 2-8 today doesn't mean you're likely to go 9-1 tomorrow, it might be 2-8 again.

[Roger T. Bannon]

It's not about discipline but about what's mathematically correct. Let's say that according to half kelly I should always bet 1%. After two weeks my BR doubles from $1k to $2k, and 1% is now 2% of the original BR. So my next bet according to half kelly should be $20, there is no reason not to bet $20. If I deposited $2k right away the bet would also be $20.
Betting more after a run of losses is gambler's fallacy. Just because you go 2-8 today doesn't mean you're likely to go 9-1 tomorrow, it might be 2-8 again.

Mathematical correctness requires precision of which you have none.

The only thing worse than having no edge and betting is having an edge and overestimating it.

The solution to your approach is readily apparent. You follow the paper version.

You cannot ever really know your edge so it is best to assume none at all. If you are going to assume an edge, make sure it underestimates.

Bet sizing is best left to the math guys that can measure it.

[Gaze73]

You do not know your edge. If you do not know your edge, you will not win.

Right, the thousands of picks in my spreadsheets must have made a profit by sheer luck. I must be due a -20% roi run over then next 1000 picks I guess.

[Roger T. Bannon]

Right, the thousands of picks in my spreadsheets must have made a profit by sheer luck. I must be due a -20% roi run over then next 1000 picks I guess.

Here again is your problem. I said you do not know your edge.

[Gaze73]

Here again is your problem. I said you do not know your edge.

If I had 5k picks at 10% roi I'd be pretty sure about my edge. I don't need 100k picks to confirm that I know what I'm doing.

[KVB]

It's not about discipline but about what's mathematically correct. Let's say that according to half kelly I should always bet 1%. After two weeks my BR doubles from $1k to $2k, and 1% is now 2% of the original BR. So my next bet according to half kelly should be $20, there is no reason not to bet $20. If I deposited $2k right away the bet would also be $20.
Betting more after a run of losses is gambler's fallacy. Just because you go 2-8 today doesn't mean you're likely to go 9-1 tomorrow, it might be 2-8 again.

This is wrong. It's a discipline problem. If it were a math problem alone, then you would outlast the variance and not even be asking this question.

But that's not what's happening. Instead, you are creating a staking problem. I can not emphasize enough, if you jiggle your bet size it will cost you a ton in vig.

You may not even be using partial Kelly correctly, if every bet you make is the same stake across the board.

Gaze, you absolutely should not be betting a percent of your current bankroll, you are not built that way and neither is your capping. You should be betting a percent of your starting bankroll and changing your bets like I said above, when up 50% or down 25%, or so.

I can give you a workable table, allowing you to bet up to 2% of your starting bankroll and can show you that you only need to increase your bets size 10 or 12 times while building a bankroll from $1000 to $100,000.

This type of betting is best for your type of gambling.

[Roger T. Bannon]

If I had 5k picks at 10% roi I'd be pretty sure about my edge. I don't need 100k picks to confirm that I know what I'm doing.

You have already confirmed that you do not know what you are doing and that is not going to change.

[Gaze73]

This is wrong. It's a discipline problem. If it were a math problem alone, then you would outlast the variance and not even be asking this question.

But that's not what's happening. Instead, you are creating a staking problem. I can not emphasize enough, if you jiggle your bet size it will cost you a ton in vig.

You may not even be using partial Kelly correctly, if every bet you make is the same stake across the board.

Gaze, you absolutely should not be betting a percent of your current bankroll, you are not built that way and neither is your capping. You should be betting a percent of your starting bankroll and changing your bets like I said above, when up 50% or down 25%, or so.

I can give you a workable table, allowing you to bet up to 2% of your starting bankroll and can show you that you only need to increase your bets size 10 or 12 times while building a bankroll from $1000 to $100,000.

This type of betting is best for your type of gambling.

You're contradicting yourself. How can jiggling the bet size cost me anything? A +EV bet is a +EV bet, the more you bet the more you win long term.

I don't even use Kelly, I do pretty much what you suggest, but it doesn't make a difference because once my stakes are 50-100% bigger than they used to be, I start losing.

Let's say I go 30-10 this week and double the bankroll. Now in the next week should I bet $10-15-20 or nothing because a bad run is imminent? If I bet $20 I'll go 10-20 so even with an overall record of 40-30 I'll break even. And betting less than $20 is scared money.

[Roger T. Bannon]

You have no chance.

[Gaze73]

You have already confirmed that you do not know what you are doing and that is not going to change.

I could outcap you in multiple major sports.

[Roger T. Bannon]

I could outcap you in multiple major sports.

It might be true. But you would never win as much as I will.

[KVB]

You're contradicting yourself. How can jiggling the bet size cost me anything?...

Here, get a pen and paper and do it for yourself...

...I’ve posted over and over again that changing you bet size is equivalent to progressive betting and it is costly.

Let’s keep it simple and get back to basics. Take a set of 21 bets; give yourself 11 wins and 10 losses. Now take any starting bankroll amount. To keep it simple, bet 5.5% to win 5% for each bet and put the wins and losses in any order you desire. After each bet, win or lose, adjust the next bet to 5.5% of the new bankroll, just like Gunshard says.

Now, if you didn’t change the size of your bets, 52.4%, or 11 wins and 10 losses would result in breaking even.

But Gunshard’s plan requires changing the size of your bets after each play, changing the breakeven point. How much it changes depends on the percentage bet, but the higher the percentage, the worse off you are. With my example, of 5.5%, the breakeven point nears 54%.

Gunshard should be thankful for discount books and vigorish, if he uses them, as his strategy raised his breakeven point, which was offset by the reduced vig.

Again, the percentage doesn’t matter; it’s easier to use 5.5% for this exercise than Gunshard’s recommended 5%, but the results are along the same lines.

Try it again and again. No matter what order you place your 11 wins and 10 losses, after those 21 bets, you always have 97.1% of your roll left...

[Gaze73]

If I had a $100k roll I'd just bet $1k on everything and withdraw $25k a month. Exponentially growing a small bankroll is harder.

[KVB]

...A +EV bet is a +EV bet, the more you bet the more you win long term...

If that's true, then why the thread?

Gaze, I am trying to help you here.

[Roger T. Bannon]

You cannot be what you wish you were. You are what you are.

[KVB]

A +EV bet is a +EV bet, the more you bet the more you win long term...

This is wrong. It's a discipline problem. If it were a math problem alone, then you would outlast the variance and not even be asking this question...

Perhaps you should reconsider your edge.

Or perhaps you should have discipline in betting.

You are pinning yourself in a corner where those are the only answers for you.

[Gaze73]

It's simple. If I were doing only flat stakes, I'd be in a nice profit because I have a nice edge. Sooner or later I must break the cycle of bad luck. I know I'll make a million one day.

[Roger T. Bannon]

It's simple. If I were doing only flat stakes, I'd be in a nice profit because I have a nice edge. Sooner or later I must break the cycle of bad luck. I know I'll make a million one day.

[KVB]

It's simple. If I were doing only flat stakes, I'd be in a nice profit because I have a nice edge. Sooner or later I must break the cycle of bad luck. I know I'll make a million one day.

It's not bad luck, it's lack of discipline.

Bad luck is noise and you wouldn't have started this thread if you staked to handle the variance, regardless of you edge.

That's the simpliciity of it.

[Roger T. Bannon]

I've seen some of your posts. You are overconfident and not very smart. Those tend to run in pairs.

[Gaze73]

I'm actually betting on horses nowadays where the odds are +100 to +2000 so the variance can be insane. I can lose 10% of BR in one hour then finish the day up 20%. In a good week I can go up 100%, then raise the stakes and lose 40% in a day, such as today. 10% of that was due to tilt but still. I don't know why tilt bets always go the wrong way. If you tilt bet on a 50/50 event somehow it will lose 80% of the time.

[Gaze73]

I've seen some of your posts. You are overconfident and not very smart. Those tend to run in pairs.

Look who's talking. The only advantage you have over me is a big roll. I'll catch up one day.

[oilcountry99]

If I had a $100k roll I'd just bet $1k on everything and withdraw $25k a month. Exponentially growing a small bankroll is harder.

I agree with this 100%. Not that my opinion matters. The more your unit is worth the less you have to win over the course of a season. 5 Units profit at $1000 units is $5000. 5 units at $10 Units is $50. Simple math. Takes money to make money.

[KVB]

...10% of that was due to tilt but still. I don't know why tilt bets always go the wrong way. If you tilt bet on a 50/50 event somehow it will lose 80% of the time.

Good work here Gaze. None of us, as bettors, can move forward or get better if we don't identify, correct, and learn from our mistakes.

That goes for all of us.

I knew there was a discipline issue somewhere, even if just 10%.

They may not always go sour, but the market, depending on the sport, will prey on your emotions. You are getting put into bad spots after winning and may not even realize your bigger bet isn't what you thought it was.

Besides, you must have won some big ones, or why else would you bet them anymore?

Don't force it, and stay patient and disciplined.

I honestly believe you can appreciate what I'm saying here.

[Roger T. Bannon]

Look who's talking. The only advantage you have over me is a big roll. I'll catch up one day.

Wrong. I will outperform you in SBR Sportsbook.

[Gaze73]

Wrong. I will outperform you in SBR Sportsbook.

I only do NBA and NHL and the soccer menu is too poor, so no dice.