I've been into racing for about a decade and I have to say I have never read/seen/heard anything that comes close to the analysis you give PM. Truly incredible.
I've been into racing for about a decade and I have to say I have never read/seen/heard anything that comes close to the analysis you give PM. Truly incredible.
Cheers JT. Too kind. It is a weird little fetish I have to admit. Whether it continues to be profitable, we shall see.
Okay, I'm curious to come at this from the other angle Peanuts - which horse(s) does your model say are about as likely to win the GN as Simon Church is to bag a hat-trick vs Reading on Saturday?
Who should I be avoiding backing at all costs, even it's a 50p e/w Granma special?
PM, you must seriously put some effort into all this - I hope you are loaded as a result.
At Cheltenham GC my millionaire boss went heavy on the winner because his millionaire dad has a house (one of many) on Lake Windermere.
They own horses, some with Sir AF, and the formula was "I have a house that resembles the horse name"
Having compiled the original database 8 years ago it's just grown each year. Takes a bit of time in February~ April and so I tend to place a reasonable aggregate wager on my book's top (usually) 5 selections (of course, only what I can afford to lose entirely) hopefully to make it all worthwhile but I structure the book relatively conservatively (all e/w and scaled stakes to give similar payoff from each selection) - so a highish stake but (relatively) low risk strategy, aiming to breakeven if I get one of the selections 2nd~4th and hopefully seeing 2 make the frame, so 100% profit if both 2nd~4th but, of course, a lot more if one wins. 4 winners and 7 profitable years in 8 GNs. The unprofitable one was because Sunnyhillboy lost by a nose and, because I'd had a biggish ante-post on Scotsirish that went south when he was sadly killed at Cheltenham, one place wasn't quite enough to breakeven (I think I was c.5% down). Without doubt it will draw a complete blank one year, maybe this one!.................but hopefully not.
Okay, I'm curious to come at this from the other angle Peanuts - which horse(s) does your model say are about as likely to win the GN as Simon Church is to bag a hat-trick vs Reading on Saturday?
Who should I be avoiding backing at all costs, even it's a 50p e/w Granma special?
Will have to get back to you in the morning Exiled, if that's OK. Need to get some shut eye :0)
I'm not back until after the race on Saturday (and the Reading game) so as I sit here on the beach in Koh Samui I can but do the housewives pick and bung an each way on Tidal Bay and Across the Bay. Silly not too. Hoping for a few 'Jammy git' texts when I get off the plane at heathrow tomorrow evening.
If anyones got a stan james account, email freebie@stanjames.com with your username & DOB quoting 'MIRROR' or 'RECORD' (or both separately) you will get a free £5 in your account within an hour
Thanks for that.
ditto
Spotted too late though... ;0(
I should think they will do them again in the next couple of days. Ive had 30 quid off them in the past week
Racing Post also calling that GN course ground Good, based on time of yesterday's Foxhunters, despite official going of GS(Gd places). It was the fastest officical time recorded since they changed the distance in 2001 but, as I said earlier, comparisons are a little tricky as they now start the clock a tad later (when they pass the starter). On my watch, started when the tape was raised, it was identical to the time of the 2004 race when the official ground description, endorsed by the RP, was Good.
Okay, I'm curious to come at this from the other angle Peanuts - which horse(s) does your model say are about as likely to win the GN as Simon Church is to bag a hat-trick vs Reading on Saturday?
Who should I be avoiding backing at all costs, even it's a 50p e/w Granma special?
Before vanity gets the better of me (as it usually does) and I give you some horses that the model thinks unlikely to win, let me explain why I don't normally do that.
1. Superstition. In 2011, I thought why not enhance returns by laying those that the market fancies but the model reckons can't win (the outsiders being typically really long prices on Betfair, it's difficult to make a decent book of lays). I think I even foolishly said on CL that one, Ballabriggs, couldn't win. When (naturally) he came home ahead of my fella Oscar Time, my net payout on the lays took a nasty slice out of my winnings from having had 2nd and 4th - I was still well ahead but I vowed to (try to) never be such an arrogant prick again and certainly never to lay GN runners.
2. Evolution of the model. Even when the model has served up the winner, which it has 4 times in 8 so far, it still shows up flaws. There will always be runners that go close (or win) or run significantly better than it had expected. I don't worry too much about runners that it fancied not making it because horses can have an off-day but those that materially exceeded its expectations are valuable to assess in order (hopefully) to improve the model for next time. If you dig deep enough you will usually find a stat reason for that outperformance - maybe a new stat. An example, Aurora's Encore (completely unfancied by the model before the race last year because of terrible season's form - hadn't made the frame in 7 races) won. Couldn't believe it watching it. This seemed like one I'd have to set aside as an "outlier" that could never be explained statistically. What then came to light was a remarkable peculiarity about that winner's chase career, shared by only one other horse in my database (Seagram, winner in 1991). He had a strong performance bias in favour of chases in the Spring (a 75% Win & near miss Strike Rate in his 8 chase runs during the 2 months from the start of the Cheltenham Festival, compared to a 6.7% Rate in his 15 chases outside that window). That of course explained his terrible season's form because much of it had been on Soft and/or without sun on his back. That poor season's form, of course, had also caused his handicap mark to be dropped gradually over the season to stand 6lbs below when he was beaten by a head the previous spring in the Scottish National. With the benefit of hindsight, he was "thrown in" by the handicapper on 10.03 and he certainly was no outlier.
3. Time. I don't dwell for too long on all of those runners that fail the big stats since I know they won't make my model's selections so I don't rate the entire field (after the race I will go back and complete the job on any unrated horses that finish within 30L of the winner, before doing my model-tweaking).
So, for all these reasons, I don't have a list of total no-hopers. However, with the HUGE CAVEAT that any horse can win the GN (as recently as 2001 only 2 of the 4 that finished did so without having fallen or unseated) and that no single stat is a "red-line" eliminator by the model, I will give you the horses that have at least one big negative stat to overcome:
- 7 year olds. The last 7y-o to win the GN was in 1940. No 7 y-o has finished <20L of the winner since (at least) 1988. There are two 7y-os running this year: Triolo d'Alene and Kruzhlinin.
- Previous Poor GN (as sole experience of course). Horses that failed to complete a circuit or exited at any time when out of touch or PU'd or finished >60L behind the winner (and had no better performance over the fences in another race) have not won a GN or finished <10L of a winner since (at least) 1988. This year, the runners that have to overcome that big negative: Tidal Bay (but watch out for him because his GN UR 3 years ago is his only F/UR in 26 chases & 14 hurdles and his other stats are so unusual that he could be a big threat), Colbert Station and Balthazar King.
- Major Jumping Course Blank - Every GN winner and those finishing <10L in all GNs since (at least) 1988 had chase form at one of Cheltenham (excl XC course), Chepstow, Newbury, Aintree (GN course), Wetherby or Leopardstown (all stiff-jumping LH tracks) or in a Class 1 or 2 chase at Punchestown (excl XC course) or Fairyhouse. This year, the runners that have to overcome that stat include: Double Seven, Big Shu, Vesper Bell, Raz De Maree.
- Big weight and lack of 3.5m+ form - No horse from 1988 onwards has carried more than 11.01 to finish 1st~3rd in the GN that didn't have solid form at a minimum of 3.5m at Cheltenham or 3.75m+ elsewhere or having run creditably in a prior GN with 11.01+. Only 5 horses carried more than 11.06 to do so (none of them winning of course) and all had form at 3.75m+ (4 of them at 4m+). This year the 11.02+ runners that fail the stat: Long Run, Rocky Creek, Triolo d'Alene (again) and Hunt Ball.
These runners mentioned are not necessarily the worst-rated runners (some could run very big races) but they have big negative stats to overcome. So, not really an answer to your question but hopefully of interest.
PS You do realise Exiled that I've just nailed Triolo d'Alene on as tomorrow's winner don't you?
TEAFORTHREE 8/1 with PP now MONBEG DUDE 12/1 CHANCE DU ROY 33/1 MOUNTAINOUS 40/1
Peanuts you can breath a sigh of relief as I will be doing Rocky Creek. Slightly off tack. I once had a punt on a horse called Tom Morgan. It was going ok, until it did a Luis Suarez on the horse in front and hence got disqualified. This is a punter who really should keep his shekels in his pocket!
If anyones got a stan james account, email freebie@stanjames.com with your username & DOB quoting 'MIRROR' or 'RECORD' (or both separately) you will get a free £5 in your account within an hour
Thanks for that.
ditto
Spotted too late though... ;0(
I should think they will do them again in the next couple of days. Ive had 30 quid off them in the past week
If anyones got a stan james account, email freebie@stanjames.com with your username & DOB quoting 'MIRROR' or 'RECORD' (or both separately) you will get a free £5 in your account within an hour
Thanks for that.
ditto
Spotted too late though... ;0(
I should think they will do them again in the next couple of days. Ive had 30 quid off them in the past week
Peanuts, what areas does 'The Rainbow Hunter' fall short on your model?
Seen the horse picked by a few people but can't see that you've mentioned it.
Sorry to be slow getting back steve, had a lot on today.
A few people fancy TRH, observing (among other things) that the horse finished (without its jockey, having UR'd at the 1st Canal Turn) alongside Auroras Encore last year. Ordinarily, Unseating or Falling in a prior GN on the first circuit (particularly at or before the Canal Turn) would be a big negative, unless there was a better run over the fences in another race. However, my model ignores his UR last time since he was hampered by another horse and it was that that largely caused the jockey to be ejected. The model rates TRH as having the potential to put in a bold show but not quite the credentials to put him in the shake up. Len is absolutely right about the 70 day break being a negative. Given that he's never won at beyond 25f stamina is unproven (especially since his only attempt at beyond 26f, other than the GN UR, was a struggling PU in last year's Bet365 [aka Whitbread]). With only 10.07 to carry that's not necessarily a negative on decent ground but it is when combined with the fact that he's never won a chase at a higher grade than Class 3 on Good ground (and only one Class 3 at a pace actually consistent with Good ground). That's a couple of moderate negatives and there are runners with plenty more and worse negatives than him. On the positive side, that was a good Listed Chase win at Donny in January (probably a career best) but he has gone up 8lbs in the handicap for that to OR144. He's had 5 chases at ORs above 136 and is yet to make the frame. Having said that, I think I'm right in saying that he had a wind op prior to his Donny win and so maybe he'll be able to show a higher level of form now. As I say to everyone, though my model's done a good job for me it has also invariably missed several horses each year running much better than it had anticipated and making the frame (even winning,..........difficult to believe I know!) So, if you fancy TRH, please don't let me put you off. Go with your fancy. He should give you a great run for your money and........who knows?
I'll be following your model, no question.
I was just interested to find out your thoughts on a horse that seems to be getting a lot of column inches.
Peanuts, what areas does 'The Rainbow Hunter' fall short on your model?
Seen the horse picked by a few people but can't see that you've mentioned it.
Sorry to be slow getting back steve, had a lot on today.
A few people fancy TRH, observing (among other things) that the horse finished (without its jockey, having UR'd at the 1st Canal Turn) alongside Auroras Encore last year. Ordinarily, Unseating or Falling in a prior GN on the first circuit (particularly at or before the Canal Turn) would be a big negative, unless there was a better run over the fences in another race. However, my model ignores his UR last time since he was hampered by another horse and it was that that largely caused the jockey to be ejected. The model rates TRH as having the potential to put in a bold show but not quite the credentials to put him in the shake up. Len is absolutely right about the 70 day break being a negative. Given that he's never won at beyond 25f stamina is unproven (especially since his only attempt at beyond 26f, other than the GN UR, was a struggling PU in last year's Bet365 [aka Whitbread]). With only 10.07 to carry that's not necessarily a negative on decent ground but it is when combined with the fact that he's never won a chase at a higher grade than Class 3 on Good ground (and only one Class 3 at a pace actually consistent with Good ground). That's a couple of moderate negatives and there are runners with plenty more and worse negatives than him. On the positive side, that was a good Listed Chase win at Donny in January (probably a career best) but he has gone up 8lbs in the handicap for that to OR144. He's had 5 chases at ORs above 136 and is yet to make the frame. Having said that, I think I'm right in saying that he had a wind op prior to his Donny win and so maybe he'll be able to show a higher level of form now. As I say to everyone, though my model's done a good job for me it has also invariably missed several horses each year running much better than it had anticipated and making the frame (even winning,..........difficult to believe I know!) So, if you fancy TRH, please don't let me put you off. Go with your fancy. He should give you a great run for your money and........who knows?
I'll be following your model, no question.
I was just interested to find out your thoughts on a horse that seems to be getting a lot of column inches.
Thanks
Possibly partly due to the fact that a part owner is a Racing UK presenter.
Is the ground to Swing Bill's liking still or too dry do we know?
On yesterday's time-based going of Good (with a nadge of cut) it may be quicker than ideal for Bill but I'd like to time the Topham this afternoon - there was supposed to be rain last night. Will let you know Len.
At lunchtime today the official ground description is unchanged on the GN course: GS (Good places) and stick reading of 6.6. They've watered between 2nd last and last fence. Will time the Topham. Maybe Warne in the Foxhunters was an exceptionally high quality winner but he would have to be about 16-20L ahead (i.e. 4-5 seconds) of the "typical" previous winner for that time to be consistent with GS(Gd).
Comments
Who should I be avoiding backing at all costs, even it's a 50p e/w Granma special?
4 winners and 7 profitable years in 8 GNs. The unprofitable one was because Sunnyhillboy lost by a nose and, because I'd had a biggish ante-post on Scotsirish that went south when he was sadly killed at Cheltenham, one place wasn't quite enough to breakeven (I think I was c.5% down).
Without doubt it will draw a complete blank one year, maybe this one!.................but hopefully not.
Silly not too.
Hoping for a few 'Jammy git' texts when I get off the plane at heathrow tomorrow evening.
1. Superstition. In 2011, I thought why not enhance returns by laying those that the market fancies but the model reckons can't win (the outsiders being typically really long prices on Betfair, it's difficult to make a decent book of lays). I think I even foolishly said on CL that one, Ballabriggs, couldn't win. When (naturally) he came home ahead of my fella Oscar Time, my net payout on the lays took a nasty slice out of my winnings from having had 2nd and 4th - I was still well ahead but I vowed to (try to) never be such an arrogant prick again and certainly never to lay GN runners.
2. Evolution of the model. Even when the model has served up the winner, which it has 4 times in 8 so far, it still shows up flaws. There will always be runners that go close (or win) or run significantly better than it had expected. I don't worry too much about runners that it fancied not making it because horses can have an off-day but those that materially exceeded its expectations are valuable to assess in order (hopefully) to improve the model for next time. If you dig deep enough you will usually find a stat reason for that outperformance - maybe a new stat. An example, Aurora's Encore (completely unfancied by the model before the race last year because of terrible season's form - hadn't made the frame in 7 races) won. Couldn't believe it watching it. This seemed like one I'd have to set aside as an "outlier" that could never be explained statistically. What then came to light was a remarkable peculiarity about that winner's chase career, shared by only one other horse in my database (Seagram, winner in 1991). He had a strong performance bias in favour of chases in the Spring (a 75% Win & near miss Strike Rate in his 8 chase runs during the 2 months from the start of the Cheltenham Festival, compared to a 6.7% Rate in his 15 chases outside that window). That of course explained his terrible season's form because much of it had been on Soft and/or without sun on his back. That poor season's form, of course, had also caused his handicap mark to be dropped gradually over the season to stand 6lbs below when he was beaten by a head the previous spring in the Scottish National. With the benefit of hindsight, he was "thrown in" by the handicapper on 10.03 and he certainly was no outlier.
3. Time. I don't dwell for too long on all of those runners that fail the big stats since I know they won't make my model's selections so I don't rate the entire field (after the race I will go back and complete the job on any unrated horses that finish within 30L of the winner, before doing my model-tweaking).
So, for all these reasons, I don't have a list of total no-hopers. However, with the HUGE CAVEAT that any horse can win the GN (as recently as 2001 only 2 of the 4 that finished did so without having fallen or unseated) and that no single stat is a "red-line" eliminator by the model, I will give you the horses that have at least one big negative stat to overcome:
- 7 year olds. The last 7y-o to win the GN was in 1940. No 7 y-o has finished <20L of the winner since (at least) 1988. There are two 7y-os running this year: Triolo d'Alene and Kruzhlinin.
- Previous Poor GN (as sole experience of course). Horses that failed to complete a circuit or exited at any time when out of touch or PU'd or finished >60L behind the winner (and had no better performance over the fences in another race) have not won a GN or finished <10L of a winner since (at least) 1988. This year, the runners that have to overcome that big negative: Tidal Bay (but watch out for him because his GN UR 3 years ago is his only F/UR in 26 chases & 14 hurdles and his other stats are so unusual that he could be a big threat), Colbert Station and Balthazar King.
- Major Jumping Course Blank - Every GN winner and those finishing <10L in all GNs since (at least) 1988 had chase form at one of Cheltenham (excl XC course), Chepstow, Newbury, Aintree (GN course), Wetherby or Leopardstown (all stiff-jumping LH tracks) or in a Class 1 or 2 chase at Punchestown (excl XC course) or Fairyhouse. This year, the runners that have to overcome that stat include: Double Seven, Big Shu, Vesper Bell, Raz De Maree.
- Big weight and lack of 3.5m+ form - No horse from 1988 onwards has carried more than 11.01 to finish 1st~3rd in the GN that didn't have solid form at a minimum of 3.5m at Cheltenham or 3.75m+ elsewhere or having run creditably in a prior GN with 11.01+. Only 5 horses carried more than 11.06 to do so (none of them winning of course) and all had form at 3.75m+ (4 of them at 4m+). This year the 11.02+ runners that fail the stat: Long Run, Rocky Creek, Triolo d'Alene (again) and Hunt Ball.
These runners mentioned are not necessarily the worst-rated runners (some could run very big races) but they have big negative stats to overcome. So, not really an answer to your question but hopefully of interest.
PS You do realise Exiled that I've just nailed Triolo d'Alene on as tomorrow's winner don't you?
Going to Paddy Power in Bromley today.
Gissus four names for a E/W bet
OK, just for you:
TEAFORTHREE 8/1 with PP now
MONBEG DUDE 12/1
CHANCE DU ROY 33/1
MOUNTAINOUS 40/1
Slightly off tack. I once had a punt on a horse called Tom Morgan. It was going ok, until it did a Luis Suarez on the horse in front and hence got disqualified. This is a punter who really should keep his shekels in his pocket!
todays code is FREEBET i'm told...
Only headlines.
Thanks, will be taking the life savings (£37.52) and putting the lot on these four.
both £10 e/w
I was just interested to find out your thoughts on a horse that seems to be getting a lot of column inches.
Thanks
Will time the Topham.
Maybe Warne in the Foxhunters was an exceptionally high quality winner but he would have to be about 16-20L ahead (i.e. 4-5 seconds) of the "typical" previous winner for that time to be consistent with GS(Gd).
Heres to the big one this year