Money Management Question

[Silver & Black]

Units - I've never really understood why someone would bet "multiple units" on a game. It's not logical to me. If you like the game, bet it. If not...stay away from it.

The money management system that I use (sometimes successfully) is 2% of my bank on every game I play, and each time I increase my bank by 20% I recalculate. EX $5000 x 2% = $100 bets. $6000 x 2% = $120 bets. $720 x 2% = $144 bets...etc etc etc.

I respect the opinions of every one on this board, please explain to me "your" money management system, and "why" betting UNITS is a good idea..

thanks

[Stifler's Mom]

I agree with you in that flat betting is the best way to win.

I occasionally add a unit or two if i really like a play, or subtract one if it's strictly a "value play" for me, but 98-99% of my bets are for the same amount.

I have thought about using the increasing % of my bankroll thing, but have never done it. Reason being that if i hit a winning streak and increase a nice amount, then bet larger amounts, and hit a losing streak, it will eat more of my bankroll taking the losing streak than I gained on the winning streak.

I do understand how it is theoretically a good idea though, it's just not for me.

As far as the units thing (say 1-5 units, based on the strength of the play), I don't get it either....alhough to each his own, and if it works for some guys, then more power to them.

[Silver & Black]

Originally Posted by Stifler's Mom

I have thought about using the increasing % of my bankroll thing, but have never done it. Reason being that if i hit a winning streak and increase a nice amount, then bet larger amounts, and hit a losing streak, it will eat more of my bankroll taking the losing streak than I gained on the winning streak.

I understand exactly what you're saying on this Stiff..I dropped 14 straight a few weeks ago, but remember, I'm only playing 2% a game. By losing 14 straight I lost 28% of my bank, but losing streak's like that are usually pretty rare.

[Jerbeek]

I agree, I have never understood the unit thing. If a play is good enough to make then it is a play. Most people do not have a better % for the higher unit plays then they do for their lower unit plays. If that is the case then you are just hoping to get lucky and have the right plays as your bigger unit plays.

As for the % of bankroll, I think if you look at it long term and put it into a spreadsheet you will see how you can really increase your profits by upping your plays as you go (considering that you are a winning capper). However, the short term can be frustrating because you will be losing the bigger plays and then dropping down and winning the smaller so it may seem like you are not as successful when looking at a limited time. I would suggest tracking your plays by straight bets and then doing a spreadsheet tracking how you would be doing if you were upping the plays based on % just to see how the difference would affect you. I think I have an excel spreadsheet that would calculate the amounts if you are interested.

[rjp]

As many can attest, I've historically been a strong proponent of flat betting. That said, over the past few weeks I've spent a lot of time trying to understand the kelly criterion, as it seems most every true professional that makes a living off sports betting uses it.

To make a long story short, the kelly criterion will grow your bankroll at the maximum rate. With sports betting, however, edges are rarely cut and dry, so a half kelly stake is the best way to go. This reduces volatility while at the same time keeping you from overbetting. It's better to underbet and than it is to overbet. Underbetting will keep you from making as much, but overbetting will cost you money.

The easiest example is when you've got a 5% edge at -200 odds versus a 5% edge at +200 odds. For the -200, you figure to have a 70% chance of winning, so you should bet 5.00% of your bankroll (half kelly; full kelly is 10%). For the +200, you figure to have a 35% chance of winning, so you should bet 1.25% of your bankroll (again, half kelly; full kelly is 2.50%).

The reason you do this is because of a concept called expected growth. For the bet you figure to win 70% of the time, the expected growth of your bankroll is 0.189864%. For the bet you figure to win 35% of the time, the expected growth of your bankroll is 0.046631%.

This is the lay discussion, as unless you want to dive into logarithms and multi-variable calculus this is the easiest way to understand it. Keep in mind the kelly criterion is a markov chain process in that the past and future sizes of your bankroll don't affect the optimal amount to bet, meaning you should always bet the % of your bankroll at it's current size. If you're placing multiple bets, then bet the ones with the highest expected growth first and then re-calculate the size of your bankroll before making new bets.

Here are some good resources if you want to read more:

Kelly criterion - Wikipedia, the free encyclopedia
Kelly Criterion Explained
Kelly Betting Index
The Kelly Criterion
Wagering Tools - Kelly Tool

I've also built a kelly criterion calculator, but won't link to it for obvious reasons.

[rjp]

I also meant to add that for the purposes of record keeping you can say that whatever your bankroll is at the start of the season is 100 units, and then just stat that you're betting X amount of units based on your kelly stake. This will fluctuate over time, of course, but it keeps things at the same starting point.

[Silver & Black]

Love the info rjp, Quick question;
correct me if I'm wrong, but if a wager is based on and adjusted to a given bankroll, how do you make multiple plays at the same time with out anticipating an outcome? I use to play a system in Football that was based on "anticipated" W/L percentage. It was the key in dictating the bankroll percentage that would be wagered. When mulitple wagers were made for games being played at the same time, there was really never a true read.

for those who do play units, please explain to my why...it's not by any means a criticism, just curious.

[rjp]

S&B, you can do it one of two ways. What I currently do is assume any pending wagers have lost. I've had really bad streaks before, so this will keep me from overbetting (again, it's better to underbet than it is to overbet). In the event I'm going to make two or more bets at the same time, I bet them in the order from highest to lowest of expected bankroll growth.

You can also do a fancy calculation where you expect to win X a % of the time while you lose Y a % of the time and then use that to get an "expected" bankroll size, but based on the research I've done and the understanding I have of the Kelly Criterion I'm more comfortable with the method I'm currently using where I assume any pending wagers have lost.

[Silver & Black]

I see how this will work, but at the same time, you're ranking your wagers, which is the only part of the MM system that I struggle with... I appreciate the information, and I will look closer into the kelly criterion. You've directed me Baseball Betting 2007 among other things, so as far as I'm concerned if you have a positive opinion...It's worth looking at..

[Silver & Black]

still waiting on an explaination as to why betting "Units" is the way to go!!

[rjp]

S&B, even if you simply use the Kelly criterion at a basic level then you're going to be betting appropriate sizes. Let's say you don't have a strong opinion on how large your edge is, but you can say that to make a bet you must have a 5% edge. These are all "ranked" the same (each has a 5% edge), but based on the likelihood of winning the bet you should be risking different amounts. Not because you feel stronger about one, but because the higher the probability of winning the faster you'll grow your bankroll.

[birdsfan5]

I like Units because it is how I started gambling and it has worked for me.

[Kevin]

Silver and Black,

You are 100% correct. Your doing it the right way and we all should follow.

I think most don't bet that way because their bankrolls are more like $100-$500 and nobody wants to play $2 to $10 bets.

Id also like to add that those (like me) who bet more units on certain games are just asking to commit financial suicide with their bankroll. I sure don't lead by example which I'm not real proud of. On the flip side, I've always been the one who runs into the fire while others are running out. I guess it's just my nature to wrestle with odds lol.

[wsox08]

units are what I started with so I've just kept with it...

[Silver & Black]

Originally Posted by Kevin

Silver and Black,

You are 100% correct. Your doing it the right way and we all should follow.

I think most don't bet that way because their bankrolls are more like $100-$500 and nobody wants to play $2 to $10 bets.

Actually Kevin, you answered my question...it's called the "Rush" Let's face it, If you started with a $500 bankroll, and played 4% on every game, that's a $20 starting bet.. BORING But I'm too old for the "adrenalin rush" that happens when you've got 20% of your bank on the Padres / Reds over 7.5 and San Diego is up 7-0 going to the top of the 9th.

As far as me doing the right thing, and "Everyone Following"... the beauty of this board is that regardless of the day, or the time of day, if you want to learn something, the ability to do so is here...at no cost to anyone.

With all due respect, and I mean this sincerely, to Bird and wsox, when I was a small child I liked strained peas and strained prunes. When I got older, I found that there was something better, so I changed.