Csf Trial

M

Melendez

At the Start
Joined
May 2, 2003
Messages
3,035
Location
Dublin
I believe, maybe incorrectly, that there is maybe some small value to be gleaned out of CSF's.

The basis of this trial will mainly be to back in races where a horse or horses prominent in the betting have a radically disproportionate chance of coming second than my idea of what the CSF will grant them.

Roughly, the selected races will fall into the following categories, though I reserve the right to add more if I think of them.

1) Odds on shots, whereby if they fail to win they will have shown such a vast disimprovement in form that, with the knowledge they have not won, you would not make them favourite to come second.

2) Horses with strong interlinking form such that with the knowledge that one of them wins you would fancy the other not to be far behind.

3)How to describe this.
Exaggerated example.
Three horse race, A, B and C.
Horse A raced against horse B 14 times over course, ground and distance. On each occasion he has beaten him by 25l. It is widely expected that horse A will again beat horse B by 25 lengths today. In truth, horse B doesn't have a snowballs chance in hell of beating Horse A. Horse C is trained by A. Mug. A. Mug does not hold out much hope for the prospects, long or short term, for horse C who is as yet unraced. He did however win a race in 1977 with a horse he thought was useless and the bookies are cognotive of this fact.

The betting used for calculating CSF
1/20 Horse A
18/1 Horse C
25/1 Horse B

Match betting BvC
1/4 Horse B
4/1 Horse C
There are probably shorter winded ways of describing this.

3) Others (to be used sparingly)

-(Front running) Top weights on lousy ground.
-Decent out of form horses dropping in grade.
-Stay or bust types moving up in trip.
-Horses with a propensity not to finish.
-Huge field Irish maiden hurdles where the CSF has to account for 14 20/1 shots who really should be 100/1 - I may chose to ignore this one for the purposes of this trial, I feel there is some value there but not in the same way as the others.

The point of this will not be to pick two horses I fancy to finish first and second. It is purely value based. It is based on the "what if you knew X would win" is there value in finding the runner up? rather than "I think X will win and Y will come second".

I will stake 6 points per race either a straight one way forecast, 2 x 3 point forecasts or 6 x 1 point forecasts. I'm not getting involved in weightings.

My aim is not just/even to make a profit, but to outperform An Capall's Bumper Bonanza.

I do not scout through every card every day so if anyone sees an opportunity please post and I may include. Critical analysis of selections or the overall trial would also be appreciated, preferably before but also after the event.

Results:

Race 1: -6.00
Race 2: -6.00 RT: -12.00
Race 3: -6.00 RT: -18.00
Race 4: -6.00 RT: -24.00
Race 5: -6.00 RT: -30.00
Race 6: -6.00 RT: -36.00
Race 7: -6.00 RT: -42.00
 
I do reverse forecasts almost every day - which of course involves finding the 2nd - and win quite a few of them. It's surprising how often the 2F wins! Mind you I don't always do F and 2F, tho I do usually include the F.

I can't follow quite what you are getting at here, but it seems a bit of a complicated way of picking out the two likeliest to come in, in front of the rest :P So many horses are on paper at least two horse races, it makes sense to RFC rather than try to bet against the short Fav, imo. It's a kind of insurance.
 
but it seems a bit of a complicated way of picking out the two likeliest to come in
Click to expand...

Absolutely not. Only in rare circumstances will it be the likeliest 2. This is another addition to the vein persuit of value. Looking for 16/1 on a "real" 12/1 shot.

A real example (that admittedly failed miserably) would be Desert Quest's first run over fences. Josear and Triggernometry looked head and shoulders above the rest of the field, but were so far behind Desert Quest that if you knew that one of them had won, you'd be pretty confident Desert Quest hadn't come second. Of course he could have jumped sufficiently like an elephant to split the pair, but he really wasn't very likely to come second. The csf would have him odds on to come second given that something else won..

As it turned out it didn't really come close, and that is the way that most of the trial will go, which is a bit of a weakness in the thing as I'm bound to get bored with it, either in success or failure, before a statistically relevant sample will emerge. There tend to be only one or two of these opportunities a week.
 
My aim is not just/even to make a profit, but to outperform An Capall's Bumper Bonanza.
Click to expand...

You are tilting at windmills my friend. After a very solid start the Bumper Bonanza is about to go nuclear.
 
It will make interesting reading. The first thing that comes to mind at this time of the year are classy horses who are poor jumpers. Short price but very much win-or-bust material.

Personally I would avoid the 'Decent out of form horses dropping in grade' as I like to make money from the very opposite, but it is all about picking them.
 
I know what you are getting at Melendez-in the past I had a couple of nice touches on csf's but i haven't tried to land one in the last 12 months or so.Something really needs to jump out at me-I think if you do the csf on a regular basis the bookies margin destroys you.
 
I agree with Luke.

Anybody who does the CSF on a regular basis is burning money.

There may be a way around losing money on the bet, and you may be able to find it, but you are essentially starting on the back foot once you decide to do a forecast.

Although having said all that, Headstrong always seems to be successful with this type of bet.
 
Hard to disagree with either Luke or BTB - essentially, I'm counting on the selectivity thing. To keep it going, I'll try to find 1 or 2 a week, although that may require adding in a few marginal ones.

I can't say I'm ultra confident, although total anhialation can prove as useful as success.
 
Mathematically, the CSF is a poor bet because it's been devised by bookmakers to produce a healthy GP % for themselves, but because it is a rigid formula (if virtually impossible to comprehend) it will throw up scenarios where it represents mathematical value and if someone has the patience to wait for these scenarios, then they should turn a profit and good luck to them. I'll watch this with interest.
 
Generally speaking, on the rare occasion I get this type of bet up, the csf pays far better than the exacta. If it is not, you have to question whether it was a smart bet in the first place, even if it does win.

I suppose I should target an end to this or it I could end up like placebacker at sheffield. I'll aim for 50 races, which isn't a great sample, but I can't see myself getting any further.
 
Thanks for that Bar, hadn't realised Scriptwriter had been sold!

There's a double for you - Scriptwriter and Kruguyrova in the nervous chase. She is a monster - was hellishly impressed with her at Kempton on her British debut. She would have won had she not jumped markedly left throughout the race - Paddy wasn't especially hard on her either. The filly looked a monster in the paddock, she's enormous and very well built, look like she would eat the lot of them that day.

Interestingly I remarked to connections that I didn't think they'd waste long over hurdles with her but was told that her likely target was the mare's final over hurdles. Trust me, this horse should win, and easily.
 
As I would have pointed out to connections had I been there, but pointed out to you later, the mares final is for novices, and Kruguyrova is not a novice, so not much point in targetting that!
 
I know you said that, smartarse - but even you admitted that you didn't know whether the new Cheltenham mares' final was for novices or not!!!!! :P
 
Oh, feck orf!!!! :laughing:

I'm just repeating what I was told!

Anyhow, just back the beast tomorrow - it is a monster, it must win.
 
Originally posted by Bar the Bull@Dec 11 2007, 08:50 PM
Can you get an angle in the 12.50 at Hexham?

Or even the 1.20, with Scriptwriter never having jumped a fence in public?
Click to expand...
I'm itching to get started but not tomorrow.

The 12.50 obviously has a big juicy odds on shot 1st time over fences on heavy ground for a 4 year old, which, on the face of it looks lke a disaster waiting to happen, so the obvious would be a reverse on the second and third favourites.

Had the favourite not been running however I would be far more keen to go against Torkingking whose trainer is out of form (even for him), achieved little in his novice chase at the end of last year and not unexpectantly made a fool of himself in the Fighting Fifth.

An alternative would be doing the favourite, with a couple of idiots, namely Open De L'Isle and John Forbes, as Bywell boy will be up with the pace throughout and if Kruguyrova does come through as expected he may break on time for a couple of experienced jumpers to plod through.

That said it is probably on the wrong side of marginal as a bet and doesn't really meet the value criteria originally stated. There probably is tiny value in an Open De L'Isle/John Forbes reverse forecast, but hell, I only have fifty goes at this, I'd need a sample of at least 500 if I start doing those sort of dreamers bets.

In short the race just doesn't feel right.

Similarly, if you knew Lease Lend had won the 1.20, how confident would you be that Scriptwriter had not beaten the rest of the field. In truth my answer would be not much more than if I didn't know Lease lend had won, so it fails the criteria.
 
Just to clarify, for the purposes of this trial, I am fully accepting that each horse's winning chance is fairly reflected in the SP. I am not doing this as a tipping exercise. What I am trying to identify is whether there are races where, given you accept a particular horse has won, would that have a radical effect on the market for determining which horse would finish second, which is not reflected properly in the csf.
 
Originally posted by Melendez@Dec 12 2007, 12:41 AM
Just to clarify, for the purposes of this trial, I am fully accepting that each horse's winning chance is fairly reflected in the SP. I am not doing this as a tipping exercise. What I am trying to identify is whether there are races where, given you accept a particular horse has won, would that have a radical effect on the market for determining which horse would finish second, which is not reflected properly in the csf.
Click to expand...
The draw in certain races is an obvious one. Whether it's down to being drawn near to the pace or on a preferred strip, the draw of the winner may well say something about the draw of the second and the csf takes no account of the draw (the tricast formula does though).

A lot of the value of the old multiple odds forecasts disappeared with the csf. The catalyst for that was a 3 horse race where 2/7 fav Little Owl ran out leaving the 5/2 2nd fav to win from the 66/1 outsider. Given the 5/2 shot won the race, what was the likeliehood the fav met with a mishap? Yet the old multiple odds formula meant the bookmakers were giving you 47/1 the 66/1 shot would finish second given the 5/2 shot had won. Nowadays the csf in such a case would pay a fraction of what ithe old multiple odds would have paid.
 
I didn't see HT's contibution before posting mine. Basically I am saying I still look for multiple odds and frequently find them available on the Tote - not to his sort of money, mind you.

An exacta is a related-outcome bet which will interest me if the odds on offer for the combination I fancy are as good as for a multiple odds bet linking separate events. I shall explain.

Bookies generally refuse doubles on related events, such as say Man Utd to win their 3rd round FA Cup match coupled with Man Utd to win the FA Cup outright. Multiple bets are only accepted on independent events, where the outcome of one event is not supposed to influence the outcome of the other(s).

An exacta bet, that horse A will win from horse B, is a related-outcome bet. On the one hand, horse A has to win, but if that part of the bet succeeds then the probability of the second part succeeding is automatically increased. Horse B's chances of finishing second are increased because one of the possible outcomes, that horse B wins the race, has been removed.

Other factors can increase the linkage between horse A and B, such as both having a good draw, having the same line of form, a fitness advantage, going preference or whatever. In other words, the assumptions used to pick the winner might equally as well serve to pick out the runner up.

The key to these bets is value, and a bit of patience waiting for the odds to work in your favour.

My way to define value is to treat the bet as a double. In other words, I multiply the win odds of horses A and B (needless to say, these win odds must themselves represent good value in the first place). For example, if the horses I fancy are both priced at 6/1, before having a bet I will require odds of 48/1 or better, 48/1 being the odds payable for two horses winning two separate races at those prices.

This is a crude enough approach, because after all the odds against horse B finishin second are reduced by horse A finishing first and I have not taken this into account. I am reluctant to do so, howeer, and will only consider it when dealing with much shorter prices, because I am aware that the Tote's takeout on exacta pools is already higher than for single bets and this is one way of counteracting this factor to some extent.

I never go for CSF bets because by my definition they don't offer this kind of value whereas exacta pools frequently will - generally only to small money, admittedly - especially if you are opposing the favourite.
 
Originally posted by Honest Tom@Dec 12 2007, 06:40 PM
the csf takes no account of the draw
Click to expand...
Is this correct? The article that I posted above says that the formula was changed.
 
Back
Top