|
JCapper Message Board General Discussion
|
| Home |
Register
Log In |
| By | Charles Town Takeout | |
| dvlander 4/16/2008 11:01:16 AM | I have nearly a years worth of race cards for CT. Their WPS takeout is 17.3%. Whenever I do an "all-horse" inquiry, the Win ROI comes back at .73 to .74. With breakage, I would expect it to be around .80. Anyone have an explanation why this ROI is so low? MNR has the same takeout and the all-horse ROI is around .80. The reason I'm wondering is that unlike most tracks, I have trouble finding a single UDM at CT with any significant volume that has any success. It always seems like I'm swimming upstream.
| |
| jeff 4/16/2008 5:58:10 PM | The returns from a flat $2.00 win place or show bet on all horses everywhere does not equal the same thing as the amount bet minus the track take - even though on the surface it seems this should be the case. I'll do my best to explain... Roi is a function of hit rate and value. Included in the set of all horses that one can possibly bet on are sets of horses that can be broken out as follows: 1. The tiniest set of horses of all: Horses that return enough money to generate a profit for the player. 2. A tiny set of horses that returns more money than the amt bet less the track take. Included in this set of horses are post time favorites and rank=1 horses for a myriad of factors. Again, they return enough money to beat the track take but not quite enough money (standing alone) to enable profitable play. 3. A tiny set of horses returning an amount that is approximately equal to the amt bet minus the track take. 4. A larger set of horses that returns far less money than the amt bet less the track take. This is the largest set of all - and why the game can be so tough to beat. The truth - and most players will never recognize this in their lifetimes - is that from a value standpoint - almost all horses (about 7 horses out of 8) stepping into a starting gate are horrible bets. Players betting horses that fall into this set can expect to lose far more than the track take. So far as swimming upstream I agree with you. I try to let the Data Window tell me which streams to swim in. Some tracks are better left alone. -jp . ~Edited by: jeff on: 4/16/2008 at: 5:58:10 PM~
| |
| busseb 4/16/2008 11:57:51 PM | Don't know how Charlestown runs, but could it be from coupled entries? This would give you a lower win % than actual because you would have fewer betting interests than horses that ran and that would reflect in the lower return. I haven't put pen to paper, but it doesn't seem like it would take too many coupled entries to significantly lower the numbers you are seeing. ElPaso
| |
| ryesteve 4/17/2008 9:03:01 AM | To my way of thinking, a low overall ROI means the races are less chaotic and the outcomes more obvious. This may seem counterintuitive (obvious outcomes = low ROI) but look at it this way: let's say every field had 10 horses, and every favorite went off at 2-1, and every favorite won. It doesn't get more predictable than that, but the overall ROI of bets placed on every horse would be -60%. That's because 9 out of every 10 bets are being placed on horses with a -100% ROI. This is an extreme example of Jeff's point that a lot of money is being bet on horses that are horrible bets, which deflates overall flat-bet ROI significantly.
| |
| DeanT 4/17/2008 9:11:20 AM | Yep, ROI will be different based on what is winning. It seems counterintuitive and goofy, and definitely not easy to get the head around - for me anyway. No luck at CT, DV? I have played a Wednesday or two. I have no data but CPACE 1 in sprints rolled when I watched. How is pacefit 1 in sprints doing, if you don't mind? Thanks for the heads up, as I was thinking of playing it, but won't.
| |
| clocker 4/17/2008 11:01:19 AM | If you are modelling CT sprints understand that with the exception of 4.5 furlong races they are all 2 turns.
| |
| dvlander 4/17/2008 11:08:54 AM | Dean, I'll try and answer your question when I get home from work tonight. I haven't done much UDM research at CT with the new JC2008 factors yet. In my prior research, none of the otherwise reliable stalwarts (i.e., CPACE, CFA, Alchemy etc.) could grab a hold at CT. For this takeout riddle, I think what gives me trouble philosophically is that the main point of a parimutuel system (at least in a win pool) is to ensure that the track take is relatively static whether a 3-5 shot wins or a 99-1 shot wins. I have to reason that if the track take is static, the overall players return must be static. Therefore, when the all-horse ROI for a very large sample is a full 10% below the publicized takeout rate, it seems a bit weird. Anyway, I appreciate everyone's efforts to explain but I still don't think it's getting through my thick skull.
| |
| DeanT 4/18/2008 10:33:55 AM | I did a bit of work at CT awhile back as I noticed on "at a glances" that speed was ruling the roost. I did find a small thing or two with speed in short sprints and pp's in routes, but I am not sure if they carry forward. I gave up since there were other tracks to play. Dont feel bad DV about the paradox with takeout. I had a professional gambler and math guy explain it to me and it sunk in about as easy as my third year calculus class. Very weird to get your head around, imo.
|
| Reply |
|
Copyright © 2018 JCapper Software back to the JCapper Message Board www.JCapper.com |