|
By:
I'll be sure to cover all those points, JPG. Thanks for the input.
|
|
By:
Dubai Duty Free World Trophy
From a Kelly point of view, it's bad business to get the favs wrong, but I'm going to take up C o r a l ' s challenge today, who sportingly go a stand out 5/2 about Triple Aspect. You can see where they're coming from because he revealed a few kinks at Sandown when beaten by Cumani's Greek import, and it's also been a good while since he last ran. On the other hand, this does not look a strong race. J J The Jet Plane could, I suppose, recover some form back on a flat track on decent ground and the 5f should be OK for him. Fallon was good on Strike The Deal last time and my analysis of this race largely depends on me being right about him being overbet. Bank = 1310 Available odds: 5/2 (C o r a l) My odds: 2/1 (33.33%) Kelly = £87.27 (say £85 win) |
|
By:
Diadem Stakes (ante-post)
Lesson In Humility, on the comeback trail after a below par effort behind War Artist at Baden-Baden, is 16/1 for the Diadem at Ascot on Sunday. I can forgive her one bad run because she's generally uncomplicated and game, and I just hope the change of stables hasn't had an enduringly negative impact (she reportedly stumbled early in the race). If she runs to her Golden Jubilee form she has a great chance of finishing in the first three, and a better than 16/1 chance of winning. Not sure if she's a definite starter yet, but the owner did say this would be her next assignment immediately after the race in Germany. Bank = £1225 Available odds: 16/1 My odds: 10/1 (9.09%) Kelly stake = £41.75 (say £20 each-way) |
|
By:
the 16s are at B e t 3 6 5.
|
|
By:
3.45 Newmarket
Audacity Of Hope seems well exposed in decent company, but Ive been quite impressed with his form since they put the tongue tie on and gave Felon the leg-up. Limitations may be confirmed today, but worth a flutter at the price in my book. Will be fascinating to see how Swilly Ferry goes also: that Donny sales race is working out a treat. Bank = 1185 Available odds: 9 (betfair) My odds: 6/1 (14.29%) Kelly (incl 5% commission) = £35.70 (say £35 win) |
|
By:
3.55 Newbury
Lutine Charlie is 14/1 for this nursery, maybe due to what on the face of it looked a disappointing effort in the role of "beaten favourite" last time out. In retrospect, though, he was beaten by some OK types in San Cassiano, Locksley Hall & Ongoodform, all of whom have run with some credit since. Although he doesn't look brilliantly treated, 14s may be a bit too big for a horse that tries hard enough and also has a bit of form on easy ground. Bank = 1150 Available odds: 14/1 (s k y , c o r a l) My odds: 10/1 (9.09%) Kelly = £29.86 (say £15 each-way) |
|
By:
Prix Thomas Byron
Only 4 runners, and while the Godolphin horse looks good, Circumvent shouldn't be 9/1. Bank = 1120 Likely odds: 9/1 My odds: 7/1 (12.5%) Kelly = 31.11 (say £30 win) |
|
By:
French markets can be deceptive, and the pleasure of Circumvent's win in Saint Cloud was tempered by his returned price of just 4.2/1 on the PMU. He was around 9/1 ten mins or so before the off, but these "early shows" are becoming extremely unreliable. On the other hand, if he'd been 4.2/1 ten mins before the off, I wouldn't have backed him at all...
New bank = £1246 |
|
By:
5.10 Nottingham
It's likely that the win over 7f at Beverley has ended Rio Cobolo's winning ways for a while, but on easy ground and at 16/1 I htink it's worth a small bet to find out whether he can still be competitive off a mark of 74. I'd marked him down for a soft ground race, but then he went and won on good to firm at Beverley, and I expected that to be end of project. But 16/1 today is bigger than I expected. Bank = 1246 Available odds: 16/1 (gen available - best odds guaranteed at H i l l) My odds: 12/1 (7.69%) Kelly stake = £23.93 (say £13 ew) |
|
By:
Bachannal
Shouldn't you be using a separate calculation for the place part of the bet? Splitting the stake is not Kelly. |
|
By:
I'm not that sophisticated. Sometimes I decide that the shape of a race merits an each-way bet, which is an entirely subjective assessment, and my approach is to split the Kelly in two equal measures.
|
|
By:
Bachannal
The place bet is an entirely separate bet from the win bet. The % chance of the horse placing will be much higher and the returned odds much lower. You should look at it to see how different your bet would be if you did that. What is the point of being all "sophisticated" about the win bet and "unsophisticated" about the place bet. |
|
By:
4.20 Nmkt
If you like the 3/1 about Bounty Box you have to be interested in the 13/2 about Sea Of Leaves. These two are fairly closely matched on a couple of previous runs this season, and while Fallon might have made the difference last time out, there shouldn't be such a gap between them in the market. This is a tougher race than their last visit to HQ with Enact in the field, a Haggas filly stepping down in trip, and Vittoria reopposing. Bank = 1220 Available odds: 13/2 (b l u e s q) My odds: 11/2 (15.58%) Kelly stake = 31.63 (say £30 win) |
|
By:
Challenge Stakes
Main Aim has been disappointing lately, but if the (plausible?) excuses given by Stoutey are correct, he'd have a better than 6/1 shot of winning what looks a fairly tame renewal today. Some of his earlier season form efforts would put him bang there. Bank = 1190 Available odds: 6/1 (Lad) My odds: 5/1 (16.67%) Kelly = 33.10 (say £35 win) |
|
By:
Champion Stakes
Ive been impressed with Mawatheeq but this of course is a massive step up for him. I was a bit worried about the ground but it seems to be riding fast enough, which is what he needs. Bank = 1155 Available odds 8.4 (betfair) My odds: 6/1 (14.29%) Kelly = 24.97 (or £25 win) |
|
By:
3.20 Longchamp: Conseil de Paris
An interesting betting heat if you fancy taking on the unbeaten & progressive Aga-owned Shahwardi at the head of the market. He was arguably a little fortunate to beat Makt last time out, and now steps up in class. The problem is that there are a few plausible alternatives: Makt himself is avaialable at 12/1 over here, and Fallon takes the ride. His overall profile, though, suggests he's still got a bit to find at this level. Cirrus des Aigles has been on the go all season but shows no sign of regressing, but maybe he's best over 10f. And Traffic Guard was impressive two runs ago when stepping up to this 12f trip, and might hasve been ridden too agressively at Chester last time out. At GB prices the one I like most is Not Just Swing, a decent Lstd & Group performer last year who has won a couple of soft races recently on the comeback trail from foot problems. He should be fresh. 5/1 isn't great value by any means but I think he's worth a small interest this morning. At PMU prices Traffic Guard is the one that could be overpriced, and is also a probable bet for me, but we need to see how the market is shaping up around 10mins before the off. Bank = 1130 Available odds: 5/1 (P o w e r) My odds: 9/2 (18.18%) Kelly = 20.52 (say £20 win) |
|
By:
Traffic Guard is around 12/1 10 mins before the off which is a bit big for me. The ground may be a bit easy for him.
Bank = 1110 Current odds: 12/1 My odds: 9/1 (10%) Kelly =- 27.75 (say £30 win) |
|
By:
2.10 Pontefract
Thought We'll Deal Again travelled well last time out in a York nursery that's worked out OK. Some reservations today (pilot 1/53 for this trainer in 2009!, the extra furlong, may want a bit more cut) suppress confidence, and so any perceived edge is taken as minimal. Bank =1080 Available odds: 10.5 (betfair) My odds: 8/1 (11.11%) Kelly (incl 5% commission) = £13.62 (say £15 win) |
|
By:
2.40 Exeter
Tor Sturgis says Cardinal James will improve this year, and while trainers' public statements should largely be ignored, this particular horse is 66/1. Bank = 1065 Available odds: 66/1 (B e t F r ed) My odds: 40/1 (2.44%) Kelly = 10.24 (say £5 e/w) |
|
By:
Perhaps going is against it today Bacchanal.......odds reflecting this presumably.
Rebel Du Maquis ....is very obvious one.....but there are a number of other horses in race that perhaps hold a chance - in no particular order... Superius Robo Kilshannig Kilshannig is one that has had benefit of run ...and the going more likely to suit....but the fav is one to beat for fitness...... Very open looking race though ...all depends how much rain falls........ GL with bet ...hope it wins for you. |
|
By:
an "educational" ride...
Bank = £1055 - there must be better ways of making a few quid... |
|
By:
bacchanal
staking is not as important as making the selections in the first place and then getting the right price. Stakingcomes third. You've been down 300 and up 550 . Seems normal. |
|
By:
bacchanal - As I really have made no effort to evaluate the "probability" on each horse winning a race, then your thread and Sandown's thread have made me look into this aspect.......and thank you for putting up thread.
All I've been doing, so far ( mainly analysing past results for my system, then devising betting strategies based on strengths of system's ratings......"value" backs and lays etc) , is crudely go by the exact position in ratings for a race ( assuming the system will generally get value odds horses in top rated that win/place ........so there might be 3 or more horses in a race of say 12 that are used to back ). If a horse is say 3rd top in ratings where 12 runners and SP odds are 14-1 it is clearly a value back etc. However, I really should be working on being more specific on what the probability is , based on my opinion on that race...............................and the KELLY thread is something I want to have a go at to see if this method of staking does improve my returns, compared to level staking say. This is a skill worth developing. So I might have a go at this , mainly on races where I believe a horse or horses have a chance to win at value odds. I will put up a separate thread tomorrow, rather than clutter up your thread ,and will try it for a few races. GL with thread and thanks again for link for calculator........ Cheers the noooooooooo |
|
By:
Cheers DFCIRONMAN, and all the very best.
|
|
By:
3.30 Bath
Small interest in Dualagi, who's been running OK recently and has some decent form at this particular track. Bank = 1055 Availabale odds: 9.2 (betfair) My odds: 7/1 (12.5%) Kelly (incl 5%) = 13.37 (say £15 win) |
|
By:
Just spotted your post ...after race over :(
Going not right today IMO......would need it soft......and still too fast............I liked the selection though!!! |
|
By:
3.05 Brighton
Arthur's Edge has looked a promising sprinter since dropped down to 6f and 9/1 looks quite generous following his win at Goodwood last time out. He has 6lb extra for that win, but we're only talking 3lb really because he was wrong at the weights. We can be reasonably sure that young Probert will be putting it all in too. Bank = 1040 Available odds: 9/1 (H i l l) My odds: 6/1 (14.29%) Kelly = 49.57 (say £50 win) |
|
By:
bacchanal
I wonder if you are so close to this kelly thing that you may not being seeing the wood for the trees. What you are doing is investing 1/21 of your bank on a chance which you estimate as a 6/1 chance. At those odds, the probability of hitting 20 or more losers is 40%. Suppose your estimate is wrong (how do you know it is right by the way?). The market is very accurate and hard to beat. If the market is right and the chance is 9/1 then the chance of you hitting 20 losers or more is 80%. Now, most LT survivors of betting would say that 1/20 of your bank is what you put on an evens chance where the chance of you hitting a losing streak of 10 is about 4%. Even if that happens, you still have half your bank left. I didn't work out these % I got them from a site if you are interested. Anyway, GL with your trial. Remember, ask yourself why you know more than the market. It is likely that your argument is no different to what the market is using. http://www.bumblebeagle.org/horsehide/winstreaks.html |
|
By:
Those % are within 100 bets by the way.
|
|
By:
If he has 20 losers in a row, he'll lose around 65% of his bank, that's entirely normal under Kelly.
|
|
By:
The chances of losing 10 in a row on evens chances is 1 in 1024, not 4%
The chances of losing 20 in a row on 6/1 shots is 4.6%, not 40% http://www.stat.tamu.edu/~west/applets/binomialdemo.html |
|
By:
*Missed the 100 bets thing, that's pretty irrelevant though, you're forgetting how much the bankroll will have grown before the streak unless the streak happens right now.
Also you seem to think that losing 60-70% of bank is a disaster, when in fact it's a normal occurence. |
|
By:
An evens chance has a probability of hitting a winning or losing streak of EXACTLY 10 in 100 trials of 0.16716912 which is 1 in 59.82 chance.
An evens chance has a chance of hitting LESS THAN 10 losers/winners of 0.97 . Therefore it has a chance of hitting a streak of 10 or more of either of .03 which is 3 in 100. |
|
By:
What's this 100 trials thing for, do it from right now or it's pointless.
You can churn out the numbers in any form you like, the simple number that all the others end up as it that you will half your bank before doubling it once in 3. Everything else is a spinoff of that result. |
|
By:
As I have tried to explain on the other thread, for someone who plays the horses as bacchanal is trying to do, 100 bets is a lifetime when you are in the middle of a losing streak. The idea that at some time in the future if you keep ploughing on it will come right is pretty hard to take.
If you are churning out 000's of bets this "psychological" distinction may not apply. In any event, its about time that you realised that not everyone bets like you but in the majority they bet like bacchanal. |
|
By:
The period of time for a clearly defined bankroll is irrelevant, you still have a higher EG with Kelly than any other method.
For the pro, then there are other issues such as expenses but this thread has a clearly defined bankroll for the experiment. You can't "chase" money just because you have fewer bets, if you lose, you lose. That's why it's called gambling and also why pro gamblers have an advantage over the casual gambler in that they have more time to hunt down advantage bets. |
|
By:
Lori
I don't wish to hijack bacchanal's thread when this debate has run its course over on my thread. Full Kelly is a high risk strategy for the backer of horses as I have tried to explain and I've no doubt that bacchanal will find this out for himself. Anyway, the debate between us should cease because neither is going to budge an inch. It is for others to judge for themselves who is right and who is wrong as far it applies to their own gambling. Goodbye and GL. |
|
By:
Sandown 22 Oct 10:40
I wonder if you are so close to this kelly thing that you may not being seeing the wood for the trees. For the long-term my mind is actually very open, but for the purposes of this trial I am committed to Kelly. Kelly may not be for me: I am conscious of its potential drawbacks, but I am also drawn to its logic. What you are doing is investing 1/21 of your bank on a chance which you estimate as a 6/1 chance. At those odds, the probability of hitting 20 or more losers is 40%. I'm staking this proportion of the bank because I think this is pretty decent value bet, which isn't to say that I'd stake the same % on future estimated 6/1 shots. I appreciate that it's still big odds against to win this particular race! Suppose your estimate is wrong (how do you know it is right by the way?). The market is very accurate and hard to beat. If the market is right and the chance is 9/1 then the chance of you hitting 20 losers or more is 80%. I don't know my estimate on this occasion is right, and I agree that the market is hard to beat. Without getting all personal I've shown a modest ability over the years (ie the all-important long-term) to win a few quid. But, and I made this clear at the start of the thread, I have not in my view made the most of this so-called edge, and I attrribute this to a conservative staking stategy. Now, most LT survivors of betting would say that 1/20 of your bank is what you put on an evens chance where the chance of you hitting a losing streak of 10 is about 4%. Even if that happens, you still have half your bank left. Fair enough, but let me ask you this - what do you stake when you make: even shot #1 a 4/5 chance even shot #2 a 1/2 chance even shot #3 a knockout 1/4 chance? I'm not saying you should apply Kelly, but logic demands that you vary your stakes. I didn't work out these % I got them from a site if you are interested. Anyway, GL with your trial. Remember, ask yourself why you know more than the market. It is likely that your argument is no different to what the market is using. http://www.bumblebeagle.org/horsehide/winstreaks.html Thanks for the advice, link & good wishes. I acknowledge your respect for the market but you surely concede it can be beaten? |
|
By:
bacchanal
The issue of whether you should bet up when you consider that you have a much better chance has been debated ad nauseum on my thread. Like you, in the past i have been able to make a few quid but there is no doubt that BF has made the market more efficient and faster to react to out-of-line prices. Whether the market is beatable or not? Its Thursday so yes. Perhaps tomorrow I might change my mind. |
|
By:
:-)
|