|
JCapper Message Board General Discussion
|
| Home |
Register
Log In |
| By | Jeff's PA post re: growing a small bankroll into a large one | |
| dvlander 12/24/2008 12:09:46 PM | If you guys haven't read Jeff's post on the PA under the topic "Playing for Profit with Small Bankroll: Part1 Starting and Wagersize", do so. It is a great, great post. Jeff, since you were able to grow your bank from very small to very large, can you briefly touch on what impact the increasing size of your wagers had on your post-time odds? In other words, were your gradually larger bets impacting the pools and depressing your payouts a bit (i.e. law of diminishing returns) and thus negatively effecting your ROI? Thx, Dale
| |
| busseb 12/24/2008 1:55:20 PM | Sorry to have to ask again, but what is the max expected losses in a row? log (???) ------------ = max expected losses in a row log (1-win%) What is the ??? Not a math major, but should probably review what I did have. Thanks! ElPaso
| |
| busseb 12/24/2008 4:45:50 PM | Now that I know I need lots of plays to get the compounding going, would it work out if I had 5 UDM's for 1 track to get 5 plays per day (as long as I doubled up when 2 UDM's pointed to the same horse) or would it be better to play 5 different tracks with 1 play at each? Seems like 1 track with enough plays would be just as good. Or would there be some sort of randomness factor that I am omitting? Any thoughts? ElPaso P.S. Jeff - your brilliance is underappreciated.
| |
| jeff 12/26/2008 11:52:50 AM | Dale, It took a while, but following a pct of bank to win approach, I eventually did reach the point where the size of my own win bets began impacting my own odds. No not at major circuits like NYRA and SOCAL where the pools are large. But certainly at tracks like FEX, EMD, HOU, MNR, SUN, and TUP where the pools aren't so large. Before deciding whether or not to start branching off into pools other than win I did some extensive R&D into which pools I should play and why. That R&D became the basis for a thread that I posted on the old JCapper msg board titled Theoretical Best Bet. Almost all of my UDMs are designed to target horses going to post at 5-1 and up. Very few of my UDMs do well with horses going to post at odds less than that. If you run any similar UDM through the Data Window you will very likely notice that the win column is almost always more profitable than place - which is almost always more profitable than show. While some of my UDMs continue to produce very nice profits in the place column, repeated R&D indicates that backing horses to win their races in pools other than the win pool is almost always universally more profitable than backing them to place. In other words once the win pool becomes saturated to the point of diminishing returns - when there is still money left over/unused from a pct of bankroll bet sizing approach - then the correct way to bet that additional money is to bet it in the exotics: exacta, double, pick-3, pick-4, etc. One "gotcha" though... Interestingly enough, when betting anything more than $10 on any one exacta combination at tracks with small pools I've had very little long term success playing exactas profitably. My thinking is that I'm probably overbetting the pool and impacting the payoffs on my own exacta combos. At tracks with small pools I've actually had quite a bit of succes betting pick-3's and doubles. Of course I've been careful not to overload the pools where I play. -jp .
| |
| jeff 12/26/2008 12:10:20 PM | Brian, It's my opinion that a player is better off following fewer circuits than he can handle as opposed to more circuits than he can handle. You're the one who has to decide where the edge of your own envelope is. I'd much rather make fewer plays if that means my own live play winds up being "clean" or mistake free as compared to making a lot of plays if that means I'm missing plays because I can't keep up with scratches and changes or too many plays happen to be going off at or near the same time preventing me from focusing on some of them so that I can make value based decisions and structure exotics tickets on them in an intelligent manner. Profits = (Edge) x (Handle) If you are making mistakes during live play (for any reason) profits can be VERY hard to come by because mistakes translate to plays where Edge is negative. My advice to anybody trying to grow a bankroll would be to start out with live play using one or two circuits a day only. Add more tracks only after you've proven to yourself that you can handle the current workload in a mistake free manner. My 2.5 cents... -jp .
| |
| dvlander 12/31/2008 10:16:19 AM | Jeff, at the smaller tracks you mentioned, what is the approximate bet size where you noticably start depressing your own win payouts. Is it like $100 units or maybe a bit more? Thx, Dale
| |
| jeff 1/1/2009 12:52:40 PM | Let's do an example. TUP on a Sunday afternoon. Average win pool is likely to end up around 12k. I've got a CPace UDM that points out 8-1 shots that win at a 13 pct clip. Here's how the win pool is likely to end up if I pass the race:
Comment: In this smallish pool betting $400 more on the horse knocks the win mutuel down from $19.20 to $14.00. That's a haricut of more than 27 percent. Q. How many of your own UDMs would be profitable in the Data Window if you reduced the average win price by 27 percent? If your UDMs are anything like mine the answer is very likely to be "Not even one." Of course I could simply bet the minimum $2.00 on every one of my UDM selections. If I did that I'd never have to worry about knocking down my own prices. But then again I'd never have to worry about managing a bankroll either. Somewhere between $2.00 and $400.00 there lies an optimum bet size - the bet size relative to the size of the pool that maximizes profits. I developed a chart based on the above UDM example. It's really driven by a simple expected value calculation based on an avg pool size of 12k, the track take of .20, and an estimated UDM strike rate of 13 pct. Here's the chart:
Notice the realtionship between bet size and roi? Of course making a similar table for a track that has a lower takeout or factoring in a rebate will change things slightly. But in this example optimal win bet size is somewhere close to $100.00. Important note: Let's say for the sake of argument the strike rate of 13 pct in the above example is based on a national average for the UDM across all tracks whenever bettable odds occur. If you study racing data you know that the odds has an effect on strike rate. The lower the odds the lower the strike rate. Why? It has to do with the wisdom of crowds. The odds contain the opinions of thousands of other bettors whose collective opinion on individual horses contains information that differs from the information I use. In other words the odds contain quite a bit of collective intelligence. But because I'm attempting to illustrate optimal bet size relative to pool size I need to point something out. There's an important exception to the collective intelligence found in the odds: With a large bankroll it's possible to overbet the pool and actually change the odds. In fact, I can actually make any horse in the race the post time favorite. If I'm dense enough to overbet the pool and dictate the race favorite - that does not mean the horse I overbet is actually the most likely winner! Just felt I needed to point that out. < G > One other note: At some of the tracks pool size varies by day of the week. For example, on Mon-Tues at TUP the win pool averages 40-45k when the TUP signal is part of intertrack simulcasting in southern California. At other times, the avg win pool at TUP might only be 12k... so it pays to know your track and the factors that drive pool size. -jp . ~Edited by: jeff on: 1/1/2009 at: 12:52:40 PM~
| |
| dvlander 1/2/2009 11:00:20 AM | Wonderful explanation Jeff. I really appreciate you taking the time to detail it out. I have done quite a bit of actual review based on UDM performance and of course I was using actual mutuel payoffs to project my bankroll growth. If I can retrieve some pool size data, your chart will really help me forecast much more accurately. I can certainly see why you take a look at the exotics once your bet size starts impacting the win payoffs. Your one comment about the favorite and your possible impact on it is fascinating. Essentially you are saying that the wisdom of the betting public on favorites is impactful unless we personally overbet the pool. One point of confusion for me. If $100 is optimum in a 12K win pool and say you are at a $400 unit level, how can you put $300 into exotics without destroying your payouts in those. I know you'll have multiple combinations but it seems like you'd be swimming upstream almost mor ethan in the win pool. Good stuff. Dale
| |
| jeff 1/2/2009 1:35:08 PM | --quote: "One point of confusion for me. If $100 is optimum in a 12K win pool and say you are at a $400 unit level, how can you put $300 into exotics without destroying your payouts in those. I know you'll have multiple combinations but it seems like you'd be swimming upstream..."--end quote I really have to concur with the swimming upstream part of that comment. The short answer is if your pct of bank based on strength of play calls for a $400 bet on a single horse at a track like Assiniboia Downs, Charles Town, Delta, or a Fairmont Park you really need to re-think what you are doing. Serious new money on individual horses at tracks with small pools essentially means overbetting the pool. WIN, PLACE, EXA, DOUBLE, PICK-X... doesn't matter which pool... you essentially have to look at pool size and limit your bets to a pct of pool size. -jp .
|
| Reply |
|
Copyright © 2018 JCapper Software back to the JCapper Message Board www.JCapper.com |