EC1
On a break
I'll guess the cross country race has high weight horses aimed solely at the prize on a regular basis
Last edited:
Phil Smith's Handicapping Of The National
- Thread starter Maruco
- Start date
I'll guess the cross country race has high weight horses aimed solely at the prize on a regular basis
Basically because it's an 'in and out the windows' race with novelty fences around a very sharp circuit, and generally contested by lower tier horses. The National is an altogether stiffer test, and simple physics dictates that weight must have a greater bearing.
Where is Phil Smith's evidence and facts that high weights need an allowance in the National? He hasn't got any and I feel it's just done to enable him to get all the horses in the handicap.
How can Many Clouds only be entitled to only a 5lb rise in the handicap after his win after having 5lb too little winning it? Auroras Encore was put up 14lbs.
Why do the journalists never pursue the matter and get some answers? It's a very important issue that needs some discussion and some facts from Smith.
How can Many Clouds only be entitled to only a 5lb rise in the handicap after his win after having 5lb too little winning it? Auroras Encore was put up 14lbs.
Why do the journalists never pursue the matter and get some answers? It's a very important issue that needs some discussion and some facts from Smith.
simmo
Senior Jockey
Interestingly enough, and possibly a first for the internet, I'm coming round to the thinking that EC and Maruco might be right.
I was taken by EC's stats, but wasn't convinced by the arbitrary choice of 11st5 as the weight cut off - largely because the "stat" that was always bandied about prior to the change was that you couldn't win the National carrying 11st or more.
So I did a bit of research. The following is the percentage of wins achieved carrying 11st or more in a variety of long distance jumps races.
Scottish National 4m1f 16%
Midlands National 4m1f 18%
Eider Chase 4m 45%
Welsh National 3m5f 24%
Whitbread 3m5f 33%
Grand National Trial 3m4f 31%
Grand National pre-op 4m4f 16%
Grand National post-op 4m4f 50%
What do I take from that? It's harder to win carrying a large weight the further you go, either by distance or by heavy ground. (I'm putting the Eider down to the class of horses - the top weights there would be carrying 10st something in the GN).
And the weights adjustment has clearly had too great an effect on the results of the Grand National.
I was taken by EC's stats, but wasn't convinced by the arbitrary choice of 11st5 as the weight cut off - largely because the "stat" that was always bandied about prior to the change was that you couldn't win the National carrying 11st or more.
So I did a bit of research. The following is the percentage of wins achieved carrying 11st or more in a variety of long distance jumps races.
Scottish National 4m1f 16%
Midlands National 4m1f 18%
Eider Chase 4m 45%
Welsh National 3m5f 24%
Whitbread 3m5f 33%
Grand National Trial 3m4f 31%
Grand National pre-op 4m4f 16%
Grand National post-op 4m4f 50%
What do I take from that? It's harder to win carrying a large weight the further you go, either by distance or by heavy ground. (I'm putting the Eider down to the class of horses - the top weights there would be carrying 10st something in the GN).
And the weights adjustment has clearly had too great an effect on the results of the Grand National.
EC1
On a break
very good post Simmo
i think its clear that the 3 main Nationals have a lot in common re higher weights.
Thats quite understandable and as you say its distance or how testing the ground is that makes it so.
If someone wants to redress it..it should apply to each race imo...but i cannot see how anyone can come up with a way of doing it without recourse to a lot of data going back years.
Another area where top weights do the opposite of this effect...win more than they should do...would also have to be looked at...most handicaps on the flat are won by horses at the top end of the handicap...surely that means that if those at further want an ease..then those at shorter distances want a greater impost..to make the whole thing a level playing field
Last edited:
Bloody hell simmo, that's a first. :lol:
I'm pretty shocked at those stats. In fact I'm stunned that the impact has been so great, given how much can go wrong over the extra trip, and the unique nature of the fences. Thanks for pulling them together.
I'm not sure how there can be any other conclusion than Phil Smith has this badly wrong.
In addition there are some surprisingly high percentages in four of the other six. 15-20% would probably be expected as a stat against for what would generally be more exposed and higher class horses horses, but 25% upwards for the others all the way up to 50% for the National itself has genuinely surprised me.
I'm pretty shocked at those stats. In fact I'm stunned that the impact has been so great, given how much can go wrong over the extra trip, and the unique nature of the fences. Thanks for pulling them together.
I'm not sure how there can be any other conclusion than Phil Smith has this badly wrong.
In addition there are some surprisingly high percentages in four of the other six. 15-20% would probably be expected as a stat against for what would generally be more exposed and higher class horses horses, but 25% upwards for the others all the way up to 50% for the National itself has genuinely surprised me.
EC1
On a break
Taking the thread of track a little
Re the flat race handicaps.. I've got two years of RSB data..so ran handicaps through it..total races 3760
In all flat race handicaps ..not amateur or apprentice...these were the results
>7.08 = 3.06%
7.08-8.00 = 5.54%
8.01-8.07 = 7.29%
8.08-9.00 = 8.34%
9.01-9.07 = 10.45%
9.08-10.0 = 11.42%
10.01-10.07 = 17.7%
a comparison of extreme in distance
5f
>7.08 = 4.69%
7.08-8.00 = 4.32%
8.01-8.07 = 6.78%
8.08-9.00 = 9.08%
9.01-9.07 = 11.31%
9.08-10.0 = 12.47%
10.01-10.07 = 14.29%
14.5f+
>7.08 = 3.48%
7.08-8.00 = 7.39%
8.01-8.07 = 9.02%
8.08-9.00 = 8.88%
9.01-9.07 = 10.69%
9.08-10.0 = 12.19%
10.01-10.07 = 16.67%
The picture that paints is that handicap upper weight isn't having enough effect of making handicaps a level playing field..it favours upper weights
It also shows that on the flat..carrying weight over further doesn't stop upper weights winning to a similar degree they do at shorter..which is also more than their fair share
Re the flat race handicaps.. I've got two years of RSB data..so ran handicaps through it..total races 3760
In all flat race handicaps ..not amateur or apprentice...these were the results
>7.08 = 3.06%
7.08-8.00 = 5.54%
8.01-8.07 = 7.29%
8.08-9.00 = 8.34%
9.01-9.07 = 10.45%
9.08-10.0 = 11.42%
10.01-10.07 = 17.7%
a comparison of extreme in distance
5f
>7.08 = 4.69%
7.08-8.00 = 4.32%
8.01-8.07 = 6.78%
8.08-9.00 = 9.08%
9.01-9.07 = 11.31%
9.08-10.0 = 12.47%
10.01-10.07 = 14.29%
14.5f+
>7.08 = 3.48%
7.08-8.00 = 7.39%
8.01-8.07 = 9.02%
8.08-9.00 = 8.88%
9.01-9.07 = 10.69%
9.08-10.0 = 12.19%
10.01-10.07 = 16.67%
The picture that paints is that handicap upper weight isn't having enough effect of making handicaps a level playing field..it favours upper weights
It also shows that on the flat..carrying weight over further doesn't stop upper weights winning to a similar degree they do at shorter..which is also more than their fair share
Last edited:
simmo
Senior Jockey
they are a wins to runners ratio.
I've got to say, this is an excellent debate. It's all about opinions and the contribution and analysis has been first class whichever side of the debate you're on.
The point I was going to make about the X-country races was weight doesn't seem to make a normal difference to the result. The non handicap race earlier in season often gives similar results to the handicap which suggests weight carrying is hardly an issue.
Reet could be right when he says the reason is its run on a sharp course and it doesn't offer real jumping test, but I would have contended that it's equally to do with the fact that very few horses actually get home so it's likely to be the same handful of horses contesting the finish.
Contrary to Desert Orchid's earlier post, previously my view was that the handicapper could possibly be having less impact by altering the handicap at Aintree than he thinks he is, but simmo's excellent analysis has blown that completely out of the water. He's clearly had a profound impact to the extent he's handed a very unfair and undeserved advantage to those at the top of the handicap.
I think Reet is probably right, but not necessarily for the right reasons. I think the ability to adapt to all of the unique jumping test the x-country course presents is important, as is the ability to last home while maintaining a decent gallop. It takes a particular type of horse to excel just as it does in the National, and that's something Phil Smith or any other handicapper will never be able to account for, and therefore he shouldn't attempt to.
The big question this raises from Alan, Simmo, and DO's posts is whether there should be a complete review of the handicapping system. Clearly there is a bias but not in the way Phil Smith is trying to approach the National. There is a compelling correlation that needs looking at, and ironically even more so on the flat.
I suspect if further work was done to analyse the relationship between distances over jumps it may be even more pronounced. Clearly there are a whole host of reasons why horses don't cross the line together, but that is precisely what a handicap is supposed to do.
The killer question then is what changes would we make to correct what looks like an imbalance and to ensure parity?
The point I was going to make about the X-country races was weight doesn't seem to make a normal difference to the result. The non handicap race earlier in season often gives similar results to the handicap which suggests weight carrying is hardly an issue.
Reet could be right when he says the reason is its run on a sharp course and it doesn't offer real jumping test, but I would have contended that it's equally to do with the fact that very few horses actually get home so it's likely to be the same handful of horses contesting the finish.
Contrary to Desert Orchid's earlier post, previously my view was that the handicapper could possibly be having less impact by altering the handicap at Aintree than he thinks he is, but simmo's excellent analysis has blown that completely out of the water. He's clearly had a profound impact to the extent he's handed a very unfair and undeserved advantage to those at the top of the handicap.
I think Reet is probably right, but not necessarily for the right reasons. I think the ability to adapt to all of the unique jumping test the x-country course presents is important, as is the ability to last home while maintaining a decent gallop. It takes a particular type of horse to excel just as it does in the National, and that's something Phil Smith or any other handicapper will never be able to account for, and therefore he shouldn't attempt to.
The big question this raises from Alan, Simmo, and DO's posts is whether there should be a complete review of the handicapping system. Clearly there is a bias but not in the way Phil Smith is trying to approach the National. There is a compelling correlation that needs looking at, and ironically even more so on the flat.
I suspect if further work was done to analyse the relationship between distances over jumps it may be even more pronounced. Clearly there are a whole host of reasons why horses don't cross the line together, but that is precisely what a handicap is supposed to do.
The killer question then is what changes would we make to correct what looks like an imbalance and to ensure parity?
Last edited:
The design of the X-Country course is such that it's impossible to maintain a 'decent gallop' in the accepted sense of the term. There are so many twists, turns, banks and obstacles close together to break the pace up that the field is never likely to reach (or maintain) the speeds they do around the largest, widest track in the country. The corollary is comparing Epsom's 5f to Ascot's 6.
simmo
Senior Jockey
I had a quick look at the percentages for the Paddy Power, Red Rum Chase and Hennessey to see if there was an inkling of that nature and they came back at roughly 35%, 40%, 50% winners respectively carrying over 11st. The Hennessey in particular made me think that the make-up of runners (winners often an unexposed second season chaser) was affecting it more than the handicap system. Also had a look at the Racing Post Chase which came back over 50%, which kind of scuppered even that thought as my feeling is that more exposed horses tend to win that one.
We could also take the view that we don't need to make any changes at all, but could simply view the imbalances that we have identified as opportunities to assist us in our gambling activites....
Diamond Geezer
Gone But Not Forgotten
- Joined
- May 2, 2003
- Messages
- 13,884
Red Rum won at Doncaster, Warwick, Wetherby, Nottingham,Teeside, Sedgefield,Ayr, Catterick,Carlisle, Newcastle as well as Aintree
Simmo, do you have the numbers of runners above 11st in those races as a benchmark? Was there a bias?
I also think we may be missing some other independent variables. eg - In x-country there would be a disproportionate amount of amateur, less skilled, jockeys - are they clustered above or below the 11st level? Similarly, if a horse is weighted above 11st is there an increased or decreased propensity for a trainer to use a claimer and is this a determinant on the statistical outcome?
EC1
On a break
yes AC..there could be lots of stuff within the stats that skew them...its a very involved game..even with databases of results we struggle to come up with definitive answers about races as a whole.and yet Phil Smith has made an important decision on a race where lots of aspects can change results..weight probably being a minor one.
The National is a very hard race to make decisions about..simply because each year there are 39 runners that lose..for a lot of different reasons. Even with the number of years we have to go at ...spotting a "trend" is very hard..in a race such as this a 3% stat could be a positive..whereas it would be a negative in other races.
If weight over distance is telling more than the handicap system is correcting then it should show on the flat as well.
A lot has been made of what a test the National is...but a flat race over longer distances is also a test because flat horses go faster..and they still are at the end of their tether when thy finish a race. You are still getting to the bottom of the horse in each code. This is something that is also being ignored by Phil Smith..every race eeks out everything a horse has in the tank. He has taken the view that only happens in the Grand National.
On the flat the difference between 5f and 14.5+ is a larger relationship between the shortest NH trip..2 miles..and its longest 4m2.5f...in some cases on the flat they travel over 4 times further than the minimum trip..whereas over sticks its only just over double. If weight at distance isn't being weighted correctly then we should also see top weights struggling to win on the flat i would have thought.
The longest Flat handicap is the Cesarewitch..you would expect top weights to struggle here if the system is wrong..its 4.5 times longer than the minimum trip...you could say an equivalent NH race compared to minimum distance would be one over 9 miles. Its also a stiff test at the end when horses are flagging.
Horses carrying 9-5+ in this race won 8 times in 40 years. That hardly looks like a disadvantage to me anyway.
I haven't yet seen any evidence that weights need amending in any race in the way Smith has changed the National..in fact looking at flat handicap results it can be argued that top weights are actually favoured...at all distances.
Last edited:
I absolutely agree with that Alan. An Cappall asks a really important question though. To be certain we'd need the % of runners above and below the benchmark to see the true correlation, particularly given we get more occurrences of compressed handicaps these days.
The truest stat is as a representation of runners in the bands you used. I suspect it wouldn't make a significant difference but without the full picture it's impossible to be sure.
I'll go away and do the x-country out of curiosity, but mainly because it's easier to do than the great work and insight coming from simmo and yourself.
The truest stat is as a representation of runners in the bands you used. I suspect it wouldn't make a significant difference but without the full picture it's impossible to be sure.
I'll go away and do the x-country out of curiosity, but mainly because it's easier to do than the great work and insight coming from simmo and yourself.
Last edited:
Eleanora Duse
Conditional
- Joined
- Jan 2, 2013
- Messages
- 888
If I'd have had all this analysis before last Saturday I'd have got near backing the winner!:lol:
Maruco's comment is absolutely right the % mean nothing if you don't have the distribution of weights. If 20% of the Cesarewitch winners won with > 9-5 but only 10% carried >9-5 it would be a very significant result.
Maruco's comment is absolutely right the % mean nothing if you don't have the distribution of weights. If 20% of the Cesarewitch winners won with > 9-5 but only 10% carried >9-5 it would be a very significant result.
Poetic licence, DG!
That list only looks impressive until you take into account he was on the go from age 2 to age 12. He was very much an Aintree specialist.
simmo
Senior Jockey
tbh, I was being lazy (and I'm not trawling through the 320 races I used to find out the answer either!)
Ok, I've done the Festival x-country. Not too hard as it's only been run 11 times
Some interesting stats:
24% of runners carried over 11st
27.3% of winners carried over 11st
28.6% of placed horses carried over 11st
56% of win and place horses carried over 11st from only 24% representation
13.7% of the represented horses carrying 11st or less won or placed
The implication is that horses carrying bigger weights overperform against the average, and when you take places into consideration the handicap appears to favour them overall. There is without doubt no disadvantage to carrying a big weight over the trip.
Some interesting stats:
24% of runners carried over 11st
27.3% of winners carried over 11st
28.6% of placed horses carried over 11st
56% of win and place horses carried over 11st from only 24% representation
13.7% of the represented horses carrying 11st or less won or placed
The implication is that horses carrying bigger weights overperform against the average, and when you take places into consideration the handicap appears to favour them overall. There is without doubt no disadvantage to carrying a big weight over the trip.
EC1
On a break
Cesarewitch runners alloted 9-5+
2014 6/33
2013 4/33
2012 2/32
2011 2/33
2010 2/32
got bored then
Last edited:
So this we know: High weighted horses have a success bias in CC at Festival, undermining Smith's belief and legacy owner anxiety in the GN. Then - we have had 20% winners from ~10% participation on the Ces, indicating a similar bias.
Truly. Weight does not matter.
Truly. Weight does not matter.