Epsom Oaks 2011

[Spook]

We all have races that we struggle in historically I suppose? Some of these are understandable, perhaps now that I’ve enjoyed some success in the Arkle, I’d probably nominate ‘the Oaks’ as my bogey race. I won it years ago with Diminuendo then never took any interest again, before returning to the fore with succession amazingly bad punts. My high water mark was Alexandrova in what still remains a landmark of how everything in a race can go wrong. My ante-post ticket had been snaffled at 8/1 on the subsequent fav and I was sitting pretty before Jim Bolger decided that Galatee was injured after all. I’ve got a feeling I got talked into Speciosa and then found some evidence that I was wrong, and switched to Time On (3 bets now). All of a sudden I realised that this was wrong too, and made a desperate attempt to chase the errors by backing Guilia who did indeed attract support on the day. This was my 4th bet!!! And clearly I was ‘out of control’. My discipline had gone and whereas I should have simply accepted the hit and moved on, I was starting to obsess a bit and chase. On arriving at Epsom I compounded this mistake by taking an unnecessarily rude approach to the less than convincing blond haired gypsy trying to sell me her ‘lucky heather’. I don’t who Heather was but perhaps I should have purchased her as well. I’d been getting wound up in a traffic jam and could have declined more politely I suppose. Suffice to say, she put a curse on me which I laughed off explaining that as Galatee was a non-runner and she was too late. In any event, I’ve got my stake back on that one because before I left home I’d had the good sense to lay Alexandrova. The misery was compounded when I got my first salutary lesson in the vagaries of splitting Goodwood’s straight and round course. A couple of sprint races had duped this apprentice speed rater into including them in his track variance with the effect that I’d penalised Rising Cross by about 12L’s. Even at this late stage I might have recovered the impending damage that was about to be unleashed upon me with a horse that was widely available at 40/1. This proved to be the final kick in the goolies as when I got back home I realised that Rising Cross should have been the top-rated.

Next year I resolved to play safe, reasoning that the race was won by something from the first 3 in the betting on 9 of the last 10 renewals and it should be easy to solve. Suffice to say Light Shift won as 4th fav. I abandoned this tactic the year later when Look Here won at 33/1.

A change of tactic is called for again, and I’ve elected to try benchmarking trials instead (and apologising publicly to the blond haired gypsy). The two horses I want to concentrate on are Wonder of Wonders versus Light Shift, and Zain Al Bolden versus Midday. I’ll deal with the Cheshire Oaks race first won by Light Shift in 2007 before going on to be the first horse to follow up at Epsom in about 30 years.

Now the method itself is one I don’t use that often, (not least of which because I’m not really sure what it tells you, anymore than I’m necessarily convinced by its soundness) but the few occasions that I have dabbled with it, it’s proved quite accurate. Now I tend to use standard times for the exercise as these do alter over time, which should in theory at least, be making some concession to improvements in science and training etc Although you can use ‘raw time’ this should discriminate in favour of the more contemporary horse.

There are only 4 races that appear on the same card as 2007 and 2011; the Lily Agnes 2yo conditions race, The Cheshire Oaks, The Chester Cup, and a 12.5F 0-95 h’cap for 3yo’s. You need to standardise them at a set distance so as to ensure that no single race contributes disproportionately to the sample by virtue of being longer. I use a mile, but it doesn’t really matter. The issue of class pars is negated because you’re dealing like with like.

2007 = -0.86, 0.00, +1.34, -1.12 = -0.64/ 4 = -0.16 secs per mile
2011 = -1.79, -1.26, +1.13, -1.44 = -3.36/ 4 = 0.84 secs per mile
0.84 – 0.16 = 0.68

The next thing to do is establish if you’ve got any notable rogues in the sample. We can estimate that 2011 was 0.68 secs slower as a base, but just need to check the spreads between the fastest and slowest in the respective years
2007 = 2.46 secs
2011 = 2.92 secs (only 2.75L’s away from 2007, and just about acceptable)

It’s normally worth checking the individual races too
Lily Agnes = 0.93 slower in 2011
Cheshire Oaks = 1.26 slower in 2011
Chester Cup = 0.21 slower in 2011
Handicap = 0.32 slower in 2011
That’s a spread of 1.05 secs between the fastest and slowest.

There’s perhaps a little bit more ‘noise’ in the sample then we’d like, but the average points to the ground being 0.68 slower in 2011, which will need giving back to Wonder of Wonders.

0.68/ 8 x 11.5 = 0.98 secs

I don’t suppose it matters whether you add this to Light Shift, or deduct it from Wonder of Wonders, the net effect is the same.
Light Shift = 2.22.99
Wonder of Wonders = 2.24.81 – 0.98 = 2.23.83

2.23.83 – 2.22.99 = 0.84 secs / 0.17 = 4.94 (or 5L’s). Light Shift would beat Wonder of Wonders then by about 5L’s meaning she’d have come 4th in the 2007 renewal. An analysis involving raw time is pretty similar. Or if dealing in standards, Wonder of Wonders = 1.81 slw – 0.98 = -0.83. Light Shift ran to standard time, so the gap is about 4.75L’s again in favour of Light Shift

So moving onto Lingfield (and this one gets messy due to changes in standard times and a few slow races to boot). In total of five races appear on the same card; both the Oaks and Derby trials, the 7F, Gp3 for 3yo fillies, a 10F 3yo Mdn and a 10F conditions race. In 2009 the trials were won by Midday and Age of Aquarius. I for one had reservations about the true distance these were run at, believing they could have been as much as 45yds longer than advertised.

2009 = -1.73, -1.34, +1.37, -0.95, +0.82 = 1.83 / 5 = -0.36
2011 = -1.64, -1.53, +1.65, -5.20, -1.53 = 8.25/ 5 = -1.65

There is an immediate problem when 5.20 secs of the 8.25 net is attributable to Audacious, and especially so when it corresponds with 0.95 from 2009 without similar ratios of discrepancy being replicated in the other races.

The 2009 range is 3.10 secs, whereas the 2011 range is 6.85 secs (not acceptable). It becomes necessary to check for rogues
Oaks trial = 0.09 secs faster in 2011
Derby trial = 0.19 secs faster in 2009
7F, Gp3, F’s = 0.28 secs faster in 2011
10F, 3yo, Mdn = 4.25 faster in 2009
10F conds = 2.35 faster in 2009

There’s only the 3 races I’d be comfortable to use, as I believe the support races in 2011 were run at a slow pace
2009 = -1.73, -1.34, +1.37, = 1.70 secs / 3 = -0.56
2011 = -1.64, -1.53, +1.65, = 1.52 secs/ 3 = -0.50

The spreads of fast and slow are now very similar, as you’d expect; 3.10 secs in 2009, and 3.29 secs in 2011. We don’t appear to have any rogues in there, but we do have a small sample with all the risks we get to reliability. The ground doesn’t appear to have massively different though, with 2009 riding about 0.06 secs slower per mile, or 0.09 at 11.5F’s

Midday = -2.49 slw – 0.09 = 2.40 slw
Zain Al Bolden = -2.36 slw
A difference of about 0.04 secs. So Zain Al Bolden would have beaten Midday by a neck

Well it’s perhaps worth remembering that this is Midday Lingfield trial winner, and not necessarily Midday the older filly that we know today. She did let’s not forget run up a sequence of 3-0 defeats against Sariska, but even so this finding needs testing further given that Age of Aquarius beat Midday by 0.56 secs, whereas Dordogne beat Zain Al Bolden by 0.16 secs, (using standard times) as I’m not convinced that Dordogne is up to Age of Aquarius’s league (any more than I ever thought he was that good). If the fourth race is introduced the spread of times is still consistent enough to use it, (3.10 against 3.29) but it only serves to speed the 2011 renewal up as it gives them 0.54 secs back by virtue of the ground being slower as opposed to the 0.06 secs that the 3 race sample would suggest was slowing up the 2009 card

2009 = -1.73, -1.34, +1.37, +0.82 = 0.88 / 4 = -0.22
2011 = -1.64, -1.53, +1.65, -1.53 = 3.05/ 4 = -0.76

It’s only when you deal in raw time that we see things take on a slightly different edge. At mile aggregates we’d get the following

2009/ 2011 Oaks trial 104.86 & 104.11 = 2011 is 0.57 secs faster
2009/ 2011 Derby trial 104.29 & 104.00 = 2011 is 0.29 secs faster
2009/ 2011 7F, Gp3, 92.11 & 91.48 = 2011 is 0.63 secs faster
0.57 + 0.29 + 0.63 = 1.49 / 3 = 0.496
2011 is 0.496 secs faster per mile

The difference in raw time between distances reflects course contours at Lingfield, at Chester they were much more even with the percentage increment in trip being much more consistent with studies that people like Mordin has published to do with sectional times. This points to the 2011 going being faster, but this isn’t shown up to the same extent when using standard times as they’ve been adjusted since in 2007.

0.496 secs / 8 = 0.062 x 11.5 = 0.71 secs (to come off Midday’s time or go onto Usain Bolts).
2.30.49 = Midday
2.29.66 + 0.71 = 2.30.37 Zain Al Bolden
2.30.49 – 2.30.37 = 0.11 secs (meaning that she’d beat Midday by half a length)

This is where the problem develops though. Midday was -2.49 slw to standard and Zain Al Bolden -2.36 slw (last time I looked, it’s not unheard of for the RP alter a time 48 hours later). A nominal difference of 0.13 secs, yet their actual race times were 0.83 different, suggesting that the Lingfield standard has been speeded up by 0.70 secs between 2009 and 2011.

If we deal in standard times off the 3 race sample the gap is a neck in favour of the Zain al Bolden, if we deal in raw time the gap is half a length. If we mix them up it gets confusing. If we use raw time differences of 0.71 and apply than to the differential in standards, then we can Midday the victor

Midday 2.49 slw – 0.71 = 1.78 slw versus Zain Al Bolden 2.36 slw (Midday wins by 3.5L’s) I’m not sure you can do this though?

I’ve never really got the concept of ‘classic trials’ with regards to the Derby and the Oaks. Everyone tells me that Dante is the best Derby trial, but in terms of testing a horses suitability for Epsom, York would be just about the least appropriate course you could imagine. Chester with its tight turns, and Lingfield with its gradients, should be better trials. What they mean of course is that the Dante is the most prestigious race and attracts the best horses. I’m not quite so sure the same applies to the Musiadora and the Oaks though. I’ve always felt that Oaks horses can source from other areas with greater frequency.

So what does all this mean? I don’t know!!! I’m tempted to draw the following tentative conclusions.

Wonder of Wonders is about 4.75L’s behind Light Shift, but with an expectation of 3L’s improvement to follow above that of normal development that O’Brien seem to be able to routinely extract from his horses on their second outing. And that Zain Al Bolden is about on a level with Midday. Throw Blue Bunting into the mix, a possible entry in Brampire who beat colts in the Ballysax, a Musiadora winner who will be known by midweek, and even some French horses who might get supplemented like Galikova, and it could be a little bit of classic this year

[broadsword]

I like the bit about the gypsy curse after that I got a bit lost

[Spook]

The only reason for including all the calculations is so people could see how I'd come to the conclusions or point out any weaknesses that I'd overlooked. I'm not totally sold on the method to be honest, but its an attempt to get a fix on a horse through others whose level of performance is now known to us.

The bottom-line is that Wonder of Wonders is about 4.75L's adrfit of Light Shift, whereas Zain al Bolden is pretty well spot on where Midday was

[Spook]

Right then, back to my Nemesis;


I watched the Segal briefing on this other day and his rationale for backing Havant (in fact I wanted to check the wind direction for the 1000 guineas, as I reckon he might have got it wrong). Tom said it was blowing from the far side to the stand side and therefore he's saying Blue Bunting benefited and that Havant was disadvantaged. Clearly a cross wind was a factor, (as it was the day previous, as it hadn't altered) but if memory serves me right, it was coming from the stands side to the far side? and if that's the case, then Blue Bunting has won despite the wind and he's picked Havant out from the favoured side under the assumption that she was disadvantaged. I'm sure I'm right, although I'm taking it from memory, but I seem to recall that Blue Bunting was one of just a couple of winners over the two days that actually came down the stands side rail?

Anyway, the other thing Tom was big on is Guineas form for the Oaks. His theory is that the Oaks is the weakest race of all the classics and lacks depth usually (I'm not sure that the St Leger isn't the weakest myself in a comparative kind of way) but I wouldn't split hairs with him otherwise. His basic theory is that the best fillies have already run in the Guineas and that you start your search there for the one that looks most likely to be suited by the extra 4F's.

There is a slight problem with his theory though; it could be based on a slightly out-dated premise. Since 2004 the Guineas hasn't really been a factor on the Oaks. In fact winners seem to route from all over the shop, and placed horses similarly route in from quite a fair spread.

wins and places

Mussiadora 2 and 6
One Thousand Guineas 2 and 3
Irish One thousand 1 and 5
Lingfield 1 and 1
Cheshire 1 and 2
Goodwood 1 and 2
Sweetenham 1 and 0
Pretty Polly 1 and 0
others 0 and 3

It's been a feature of my own figures that for the past few years now, the one thousand guineas dominates the Oaks trials in a way that the two thousand guineas doesn't quite the Derby trials, and I probably need to find a way of discounting this a bit (hence dabbling about with benchmarking - want to try doing Beatrice Aurora against Snow Fairy), but this is where I am now with the fillies now at 9.0

Blue Bunting = 98.52 (Guineas)
Together = 97.77 (Guineas)
(Galikova = 96.48)
Emiyna = 93.58
Havant = 92.77 (Guineas)
Beatrice Aurora = 92.05 (Lupe)
Zain Al Bolden = 90.83 (Lingfield)
Spin = 86.99
Mesariya = 84.29
Wonder of Wonders = 82.83 (Cheshire)
Look At Me = 82.38
Izzy Top = 82.05 (Sweetenham)
Dancing Rain = 81.95 (Sweetenham)
Misty for Me = 80.52 (Guineas)
Sirens Song = 79.04

It's pretty meaningless as a set of ratings, I'd accept that. I've twice clocked Arizona Jewel in the low 70's this term and don't believe she's a threat despite the prominence of the Musaidora. We've got the Irish Guineas to run of course, and there is evidence to suggest this could throw up a credible challenger, but at the moment I'm not really that much nearer knowing the answer.

There's no money down yet, and although I quite liked Zain Al Bolden, I'm going off her a bit. I'm actually interested in the Lupe (or whatever its called these days).

[Spook]

Now i should say at the outset that I'm no royalist, but even I know that there's been a wedding recently, and that some mare going under the curious title of 'princess' called Beatrice, didn't distinguish herself in the area of sartorial elegance and wore some hat which looked like a pretzel. So when a horse turns up called Beatrice Aurore and wins the 'Height of fashion stakes' at Goodwood, you kind of know that its fated to win the Oaks with the Queen winning the Derby!!!

Leaving that awful prospect aside, I did do the same benchmark comparison with the 2010 card as it features identical races albeit one of the 0-80's is a 0-85 today, otherwise everything is the same.

The 2010 card was fractionally faster in 5 of the 7 races. The differences are remarkably close though. The only race I was really uncomfortable with was the 7F, 3yo, Mdn, for fillies where the difference in time was 1.99 secs and possibly the 0-65 sprint where it was 1.52 secs. Otherwise 2010 was faster by 0.57, 0.89, 0.45, and 2011 faster by 0.67 and 0.44 at mile aggregates to standard.

If I omit these two races, 2010 adjusts 0.16 secs faster and therefore 0.20 secs at race distance, which means that Snow Fairy would have beaten Beatrice Aurore by 0.37 secs (or 2.25L's). If I take all 7 races, then 2010 would be 0.61 secs faster, and Beatrice Aurore would beat Snow Fairy now by 0.19 secs (a fraction over a length).

The one that might be worth a mention actually is the 6F, 2yo, fillies Mdn won in 2010 by Memory in a time 4.42 slw and won today by Glee in a time 3.92 slw. Sprints tend to be more reliably run for pace (the only other race where the 2011 time was faster was the 2 miler), so Glee's performance in beating the Memory by 0.5 secs is potentially quite notable, and marks her out for the Albany stakes

Not that I'm any nearer solving the Oaks

[Spook]

And whilst on the subject of the Oaks, here is Nick Mordin's view (I'm a little bit baffled by it in truth) and am yet to be convinced this sectional analysis of his for slow run races actually works, but in the spirit of open mindedness;

DANCING RAIN SHOULD BE OAKS FAVOURITE

The clock says that the Swettenham Stud Fillies Trial at Newbury was that rarest of races; a Classic trial with two genuinely top class horses.

Nowadays there are so many Classic trials that most don't attract any proper Group 1 horses, and when they do it's just the one. But this race was an exception.

I concede I reached this conclusion with some difficulty due to the huge rail movements made at Newbury. They reportedly added 32 metres to the distance of races run around the turn.

Unfortunately this does not seem to be correct as it suggests the Fillies Trial was a significantly slower race than the mile and a half handicap won by Peintre D'Argent.

In fact the first ten furlongs of the handicap was run two seconds slower than the ten furlongs of the Fillies Trial and the last ten furlongs 3.4 seconds slower despite the slow early pace of the Listed race.

There are two ways I can see to fairly adjust for the slow early pace in the Fillies Trial to figure out the true merit of the race. The first is to use my sectional timing formula to factor in the speed they went over the last three furlongs. The second is to take those times from the ten furlong starter's stand at face value, taking off 1.7 seconds from the time of the handicap in both cases to adjust for the longer distance. The second approach makes the assumption that the effects of tiring over the last two furlongs in the handicap cancels out the effects of the slow early pace in the Fillies Trial. Experience tells me that my sectional timing formula produces a more reliable answer. But in this instance they both give the same big rating of 39.

The sectional times demonstrate the remarkable phenomena I've noted before of how top class horses can recover from a slow early pace to clock a decent final time. They covered the first half of the race 1.5 seconds slower than they did in the handicap but proceeded to go 4.8 seconds faster over the final five furlongs.

DANCING RAIN (39) wasted some energy by pulling hard against the slow early pace but looked sure to win when taking the lead approaching the two furlong marker. But she was somewhat marooned out in the centre of the course and understandably drifted right towards the stands side running rail. This allowed IZZI TOP (39) to go on after she'd quickened up to challenge. Dancing Rain picked up and closed the gap when she straightened up in the last hundred yards but was still a head shy of success at the line.

Both fillies were traveling really strongly crossing the line, and it was impressive how they rapidly opened up a three length gap on their rivals as they dueled through the final furlong

I suggested after her last run that Dancing Rain was the one to beat in the Oaks. And until this race no Oaks candidate had bettered the rating of 38 I awarded her that day. Now she's gone and run a bit faster.

Dancing Rain runs like she'll adapt to Epsom and stay the mile and a half of the Oaks, so she's an obvious candidate for Epsom.

Izzi Top is a little more problematic as trainer John Gosden says she didn't handle the dip at Newmarket on her previous start which raises clear doubts about the steep gradients at Epsom. In addition she looks a pacier sort than Dancing Rain. The fact that she's still in the Coronation Stakes over a mile suggests her connections have doubts about her going longer.

My feeling is that ten furlongs is Izzi Top's best distance. And her owner agrees. He said after the race "we're not sure she'll stay one mile four furlongs so it's doubtful she'll go for the Oaks."

This leaves Dancing Rain as the Oaks candidate to take from the race. She's now earned the two top speed ratings I've given all the Oaks entrants. So I think the general 25-1 you can get about her for the Oaks is plain crazy. As I see it she should be favourite.

[andthereoff]

Nick Mordin has been on the ball in recent weeks, I put up Theysken's Theory last year as a genuine 1000 Guineas shot although missing the race she won well at York which Nick has also picked up on. The horse in question Dancing Rain, well I did infact pick up on her Newbury win on the 15th of April as pretty decent although took Sunday Bess and Highest out of the race because I thought she was flattered by making all and couldn't do that again in top top races.

I think Nick is very much on the right lines here, although I do question how many gears she has which I couldn't question with John Dunlops winner today Beatrice Auroa whose been stunning in both races to date and put to bed an improving Highest more convincingly than Dancing Rain.

Beatrice Auroa!? hmm.. tempted.

Will have a word as I'm popping down John Dunlops open day next week.

[Spook]

Originally Posted by Spook

I watched the Segal briefing on this other day and his rationale for backing Havant (in fact I wanted to check the wind direction for the 1000 guineas, as I reckon he might have got it wrong). Tom said it was blowing from the far side to the stand side and therefore he's saying Blue Bunting benefited and that Havant was disadvantaged. Clearly a cross wind was a factor, (as it was the day previous, as it hadn't altered) but if memory serves me right, it was coming from the stands side to the far side? and if that's the case, then Blue Bunting has won despite the wind and he's picked Havant out from the favoured side under the assumption that she was disadvantaged. I'm sure I'm right, although I'm taking it from memory, but I seem to recall that Blue Bunting was one of just a couple of winners over the two days that actually came down the stands side rail?

Yep, the wind was an easterly on the Guineas weekend which would have inconvenienced those drawn high now that they've switched the stall numbers round. The wind that Segal is describing is a westerly. Basically he's got 180 degrees the wrong way round I reckon, and his conclusions about the respective merits of Blue Bunting and Havant based on wind direction are consequently flawed. He should be using it as evidence to support Blue Bunting not crab her

[Spook]

Originally Posted by Spook

Now i should say at the outset that I'm no royalist,

Well like I said, I kind of take a bit of pride in not being a Royalist, but I certainly defer to Willie Mullins if this quote from the RP is correct (couldn't resist it)

"Champion trainer Willie Mullins, who saddled four winners at Cheltenham, commented on her style and knowledge and added that he had recently learnt about her heritage via Florida Pearl.

"When I dug out Florida Pearl's winners rug from when he won the King George I realized that it was the King George VI Chase," he said. "Luckily, there are men with more knowledge than me on her lineage and I now know he was Queen Elizabeth's father. We certainly had a talk about that."

[andthereoff]

Does anyone know if Bible Belt for Jessica Harrington is planning on running?

[Spook]

Khawlah would be the one you'd have to be interested in if turning up (seems unlikely though)

[andthereoff]

I've come to the conclusion that my Oaks winner is coming from David Wachmans yard in CRYSANTHEMUM - Following a succesion of analysis.

Basically I've used a process of elimination through strict standards, firstly I've narrowed the field down using the the average speed rating of all winning Oaks candidates leading up to their run which came out as (86), I've used the data to project a serious candidate award using the upper quartile of that data to (89)

I've then added all the winning Oaks horses speed ratings together to create an "experience" collum and using the lower quartile through the set of results I've come up with the figure of (174)

Oaks Candidates;

- Blue Bunting
- Togetther
- Laughing Lashes
- Siren's Song

Serious Candidates

- Crysanthemum



Each month also has a set of averages;

Apr - 0
May - 0
Jun - 78
Jul - 79
Aug - 85
Sep - 89
Oct - 87
Nov - 84
/
Mar -
Apri - 89
May - 90

The two year old form probably doesn't resemble much worth after the Oaks now although I would like to use it to see fillies that have used it to show genuine progressive traits i.e 79>88>94 etc although I'll be more interest in those that have beaten the standards for April and May, they are;

- Blue Bunting
- Havant
- Misty For Me
- Zain Al Boldan
- Together
- Siren's Song


Using a matrix to calculate how each Oaks winner performed in relation to the months standard I've averaged up their score then using the middle section between the lower and higher quartile range I believe the Oaks winner can be trapped between -2.2 and 1.0, that leaves us with

-10.8 Spin
-9.7 Beatrice Aurore
-8.7 Charleston Lady
-7.7 Zain Al Boldan
-7.7 Izzi Top
-6.7 Look At Me
-6.0 Khawlah
-4.3 Havant
-3.5 Bible Belt
-3.3 Dancing Rain
-1.8 Misty For Me
-1.5 Wonder of Wonders
-0.6 Blue Bunting
0.0 Chrysanthemum
3.0 Siren's Song
3.8 Laughing Lashes
3.8 Together

I think that brings in the right horses to be fare.


/

My final say on the matter is Wonder of Wonders and Blue Bunting are the likely stayers with Chrysantheum and Misty For Me probably the value bets although I'm siding with Wachman on this one


1st - Blue Bunting (2/1)
2nd - Chrysanthemum (25/1)
3rd - Misty For Me (7/1)
4th - Wonder of Wonders (4/1)


I've attached the document I've been working ffrom if anyone finds it useful.


EDIT (30.05.11)

Chrysanthemum N/R

1st - Blue Bunting (2/1)
2nd -Wonder of Wonders (4/1)
3rd - Misty For Me (7/1)
4th - Dancing Rain
5th - Siren's Song
6th - Havant
7th - Khawlah
8th - Look At Me
9th - Izzi Top
10th - Zain Al boldan
11th - Charleston Lad
12th - Beatrice Aurore
13th - Spin

[Spook]

Trying to reverse engineer what you've done isn't easy, but I can see the logic you're trying to apply. However, there are a couple of things that should be ringing alarm bells.

In statistical terms Eswarah is what we call an 'outlier' and should be omitted from your sample as she's skewing it at -8.5

I'm not sure I can see the value of your 'experience index' ? what's this contributing to the calculation? I don't understand what column u25 to u40 is either?

[andthereoff]

Originally Posted by Spook

Trying to reverse engineer what you've done isn't easy, but I can see the logic you're trying to apply. However, there are a couple of things that should be ringing alarm bells. In statistical terms Eswarah is what we call an 'outlier' and should be omitted from your sample as she's skewing it at -8.5 I'm not sure I can see the value of your 'experience index' ? what's this contributing to the calculation? I don't understand what column u25 to u40 is either?

Thanks for that.

Don't worry about column u25 to u40 because I have been using the excel file to make notes and carry out further experiements so you'll find a lot of rubbish in there as I like to work with everything in one file.

Lets hope the final outcome isn't rubbish!

/

You had any further thoughts on the Oaks yourself?

[Spook]

No

It's a race I've always struggled with for no apparent reason that's ever been clear to me. Misty for Me threw a spanner in the works. I held her in high regard as a 2yo but had dismissed her after Newmarket, but all of a sudden she's right back in the mix. I think it could be a very good renewal actually

[Frankel]

I backed Chrysanthemum some months ago at 33s. She has always been held in high regard by David Wachman and her figures posted on her two runs last season entitle her to go very close in an Oaks, (and her run in the Guineas confirmed that view staying on late).

[Spook]

Continuing my futile 'benchmarking' exercise in an attempt to try and finally get this race won, I've turned to the Curragh and the re-appearance of Misty For Me. In this case I'm trying to 'fix her' through Peeping Fawn 2007 (won by Finscael Beo) and 2003 won by Yesterday.

Most of the races transfer, albeit there's an mile handicap in 2007 and 2003 that doesn't appear in 2011, and in 2003 the 3yo Gp3 was a listed race, but the 0-80 was a 0-90

The comparison between 2003 and 2011 is the easier one to make. All of the times were slower in 2003. The spread of differentials at mile aggregates to standard is as follows

0.55
2.02
3.88
4.54
4.90
6.21

This is too volatile to be reliable. I've chosen instead to use the median cluster of three separated by 1.02 secs and averaged the correction as being 4.44 secs slower. This is thus added to Yesterday's performance;

4.44 + -3.80 = +0.60

Misty For me was fast by +1.10 and so enjoys theoretical 0.50 secs advantage over the 2003 Oaks runner up and would have beaten Casual Look by 0.45 secs (about 2.75L's)

The 2011 and 2007 renewals shows up two clusters of time differentials. All of the 2007 times are slower apart from the 0-105 which is duly omitted. Otherwise we get a cluster of differential of 3.40 to 2.94 and 2.00 to 0.92. Perhaps significantly these two clusters also represent trip with the slower times at the shorter distances. This could be due to unfavourable conditions in 2007, or as seems more likely, favourable conditions in 2011. The conclusion I drew was that more time has to be given back to the 2007 cohort. The same thing doesn't show up in 2003, which is a minor concern, but possibly explainable.

Using distances up to a mile then, the average between the two was 3.26 secs slower, which means that Finscael Beo would adjust to 0.96 fast, which is 0.14 slower than Misty for Me. The horse I'm interested in though is Peeping Fawn who finished third, some 2.25L's away. Her time would adjust +0.57 which means Misty for Me would beat her by 0.53 secs (3L's). This in turn would be enough to beat Light Shift by 2.5L's if transferring to Epsom.

Misty for Me was a late developing filly and would go on to reverse with Light Shift and impose superiority by the end of the season

So far I've got some moderately contradictory figures, but they are based on respective trial performances and then projected onto Epsom (the latter being a leap of faith)

Beatrice Aurora bts Snow Fairy by 0.19 secs (at Goodwood)
Wonder of Wonders loses to Light Shift by 0.84 secs (at Chester)
Zain al Bolden might beat Midday by a neck (at Lingfield)
Misty for Me beats Peeping Fawn by 0.53 secs, and Yesterday by 0.50 secs (at the Curragh)

Light Shift kind of ties things together a little bit as she beat Peeping Fawn, by a quarter of a length for a theoretical trials result projected onto Epsom of; (I think)

1st - Misty for Me
2.75 - Light Shift
0.25 - Peeping Fawn
4.5 - Wonder of Wonders

I think I'm starting to side with Misty for Me, but reckon Beatrice good surprise a few and get nearer than is imagined. I suppose I really ought to try and tie Blue Bunting into this line of enquiry before the end of the day though

[broadsword]

Originally Posted by andthereoff

I think Nick is very much on the right lines here, although I do question how many gears she has which I couldn't question with John Dunlops winner today Beatrice Auroa whose been stunning in both races to date and put to bed an improving Highest more convincingly than Dancing Rain. Beatrice Auroa!? hmm.. tempted. Will have a word as I'm popping down John Dunlops open day next week.

I saw this win and was quite taken with it thought I'd have a little e/w for interest did you go down to Dunlops ? Whats the craic,run or not ?

[Spook]

I wanted to complete the set by trying to 'fix' Blue Bunting but have struggled in truth.

Kazzia is the obvious candidate on paper to compare her to, but there's only 4 races that appear on the same card, and 2 of these are run under slightly different conditions. In addition Kazzia's Guineas is a bit of a weird one and provides a rocky foundation to start with. The same method would indicate that Blue Bunting would beat her by 30L's (which clearly isn't happening). It points to something being amiss, and if it's a simple arithmetic error on my part, I can't for the life of me see it (often the case if its starring right at you mind).

I've kind of being reduced to using Virginia Waters but her Oaks performance is hardly an ideal benchmark given that she didn't stay the final furlong. I've got Blue Bunting beating Virginia Waters in the Guineas between 1.75L's and 3.75L's, but she lost to Eswarah by 5.5L's herself (not that this means that Blue Bunting would of course).

Every time I watch the Guineas I grow more impressed with the way Blue Bunting saw her race out.

I'm going to reiterate it again as I'm sure we'll hear it again from the RP in the lead up to Friday, but I think Tom Segal is wrong. He say's that it was the horses on the far side that took the brunt of the cross-wind and is using this as evidence to support Havant. I've checked the results and stuff I archived on here, (I remember mucking round with the google earth even) and I'm sure the wind on the Guineas Sunday was an easterly, which means that Blue Bunting had the worst of the conditions and still prevailed?

Might be useful if someone could check this out? as it could be really important yet for those people backing horses further back if they're under an alternative impression.

I'm actually thinking there's a 1-2-3 in play involving Blue Bunting, Misty for Me and Beatrice's Aurora

[The Duke]

Originally Posted by Spook

I'm going to reiterate it again as I'm sure we'll hear it again from the RP in the lead up to Friday, but I think Tom Segal is wrong. He say's that it was the horses on the far side that took the brunt of the cross-wind and is using this as evidence to support Havant. I've checked the results and stuff I archived on here, (I remember mucking round with the google earth even) and I'm sure the wind on the Guineas Sunday was an easterly, which means that Blue Bunting had the worst of the conditions and still prevailed? Might be useful if someone could check this out?

You would be correct. Blue Bunting was disadvantaged coming up that rail, the wind was blowing across her thus giving those down the centre of the track more shelter from the wind.

[Spook]

Thanks Jamie.

I was sure Segal had this the wrong way round, and his using it to tip Havant looks flawed therefore. Obviously the Guineas form was franked at the Curragh, but I don't think this owes anything to the cross-wind at Newmarket. It points to Blue Bunting actually being a worthy fav

[andthereoff]

Originally Posted by broadsword

I saw this win and was quite taken with it thought I'd have a little e/w for interest did you go down to Dunlops ? Whats the craic,run or not ?

They caudened her off for obvious reasons with the Oaks only a few days away, she's well but I don't think she'll be winning anyone over in the parade ring!

[Spook]

I think its a damned good renewal this one. The only thing missing really is a Mussiadora representative..... I suppose you could add a French rep to the list of absentees. You've got two Guineas winners, plus 4 good looking trial winners

[The Duke]

It's a very good renewal indeed Spook, and has the potential to be one of the best for quite a while too.

I make it between Blue Bunting, Misty For Me and Beatrice Aurore at this current stage, and I've played a few shekels on the Dunlop filly each-way (I also hold a 33/1 antepost voucher for Chrysanthemum).

[Spook]

I'm having a touch of 2005 revisited where I talked myself in and out of the entire field, backed half of them, laid Alexandrova and lost a packet. Luckily I haven't had a dime on so far so it isn't quite an action replay, but anyone asking me seems to get a different answer by the day at the moment.

Those are the three I've got it down to as well, but am swinging back to the fav now who I can't really find a fault with