Simple Simon's Overlay spotting method

[crash]

How often do we hear the claim: 'I just back the 'overlays' and easily stay ahead of the game'. Of course the claimant never goes on to tell you exactly how they 'spot' these overlays, that seem to be all over the race meeting but for their [shrewd] eyes only !!

Deciding if a runner is overvalue has very little science in it and has more to do with art and experience of a lot of practise creating your own betting line on the 3/5 best chances in a race and then only excepting prices above your own price for any particular runner. How do punters who never do a price line, spot a 'genuine' overlay ? Gut feeling ?

Without going on here to describing just how to do your own price line [and then expecting you to practise it for six months before betting with it], below is a very simple method for having a good idea as to what is good or very poor value odds offered for any runner.

Don't be surprised at just how few overlays are actually present on any given race day [especially Sat.], but yes they do exist but just not in the numbers that many punters would love to believe.

Method:
Work out the top 5 fancied runners for the race you are interested in [the top 5 in the p/post market from your newspaper is as good as any other method or pick your own method]. Now if you are a win bet man, have a look at each runner's win SR. If it is close to 33% that means a 1 in 3 win rate. Any price over $4 becomes value. If the SR is 25% any price above $5 becomes value. Get the idea ? It is simplicity itself. The same methodology can be applied for those interested in the place bet by looking at the place SR.

The above method might sound too simple, but lets remember here that any runner's 'actual' odds of winning a race can never be more than an educated guess and the simple method I have described at least gives a very good idea of value, and gets any punter thinking about value [very important] rather than 'gut feeling' it before placing a bet. This method probably has as much, if not more real relevance to price than any other method, as at least the runner's SR is set in stone. Don't worry too much about zeroing in to bet one particular runner [you like], but think of the runner offering the best price, as the object of the game is not to win every battle but to win the war !!

Good Punting.

[partypooper]

MMMM,interesting, Never thought of that angle Crash, even though I've mucked around with S/R B4, will do some work on that one.!

[crash]

Hi Party,

As a systems man I thought you might be interested, as there is a definite mechanical potential here.

The method hit me in a blinding flash when I was at work on yet another price line for a couple of races today and was considering the various SR's of some runners and thinking ...'theres has got to be an easier way' !!

When you think about it for a minute and get past the [monstrous] simplicity, well why shouldn't price be based on SR ? Of course I [or anyone else] can think of situations where the SR is misleading, such as older horses moving into twilight performance, or a runner being out of it's winning class, weight or distance etc., but looking at the overall picture of the majority of it's runs where the war is won, the method definitely has 'truth' going for it as far as a runners real odds go. Odds that are staring right at us, in the 'facts' of it's SR.

[Mark]

Crash

What do you do when you have these numbers?

Lad Of the Manor 69%
Not A Single Doubt 57%
Danehill Express 40%
Regal Roller 34%
Casual Pass 33%
Elvstroem 30%
Savabeel 30% etc

SR7 tomorrow has 5 runners with more than 40% strike rate including one at 100%.

[crash]

Mark,

I know you know how to convert % to odds so I am concluding you are asking me how to go about choosing a winner from the race's runners and if so, you are missing my whole point entirely. The method is a pricing mechanism, not a selection process.

From that race tomorrow especially [and the Aust. Stakes in Melb.], don't expect any overlays from the 2 or 3 runners who you select that you think have a good winning chance.

PS. If I can get $5 Savabeel I'll take it !!

[Lin]

Thanks for the idea, I'll follow it up. I've long been aware of the concept of
value betting, as opposed to plonking your money on the horse most likely
to win, regardless of the odds.
A problem I've had over the years is the shift in prices between the morning
paper, and the start prices. But I'm not the type to go along to races.
I wanted to ask you what you think of the idea of giving a 'weighting' to a
win result last week as opposed to a win of four months ago. ie, if a horse won two of it's last six races for a 33% SR, but the two wins were 3-4 months
ago, and it's more recent starts were disasters, would we give that horse the
same credibility as if it's two wins were the most recent starts?
I don't know a lot about the horse racing game, and tend to have trouble
remembering stats, so I hone in on simple systems like yours. In a horse race there are too many things to go wrong in a given race, you can only play the
averages game, accept losses, and concentrate on Strike Rate x odds, to
stay ahead of the game.
all the best
Lin in Gosford
PS Don't let my name fool you, I"m a bloke.

[crash]

Don't worry Lin, my real name has three L's and two B's in it [it's French] and is often mistaken for my surname. Apart from that there are some very sharp female punters around !!

You are on the right track in my humble opinion in looking for 'value' rather than winners per se.

The scenario you describe is not to be worried about, because if you choose to only look at the 5 runners [where 80% of all winners come from] with the shortest pre/post odds [or the 5 best ratings], no horse like the one you describe would be among them as the runners shocking performance would be taken into account for you and reflected in it's pre/post odds which would also be shocking.

[Bhagwan]

Hi Crash,
I like the idea ,it`s as good as any.
An easier way of calculating the odds instead of converting to percentages & then to a odds range , would be simply divide it`s number of wins into it`s number of starts e.g. 10 career starts for 1 win =10/1 or $11.00

THE PROBLEM ONE WILL FIND IS TRYING TO PLACE A BET THAT WINNS ENOUGH TO MAKE A PROFIT.

I feel a stronger approach would be to target its performance wins for
Good tracks only. e.g. 5 Good track starts for 1 win =5/1 or $6.00

That way we are comparing Apples with Apples as far as career track condition is concerend. Do this regardless what todays track condition is.
I feel it makes the theoretical price lower & more accurate , also more action.

I also feel it would be a strong idea to target only races where the whole field has had a min. of 5 career starts , that way we have a bit more meat to work with.

Cheers.

[blocka]

I think what Mark may have been getting at is that the priceline can easily add up to more than a 100% market.

For example if you have 3 horses in a race. Each with a strike rate of 50%, therefore by your method the fair price of each is $2 or even money. So then if you could get odds of $2.5 for each runner (120% market) you would then bet each as by your method they are value. Yet clearly they are not.

Therefore you need some way of making the percentages add up to 100%.

Cheers,
Blocka.

[crash]

I agree Bagman, for runners with only a few starts, the SR figures are obviously fairly meaningless as a value guide. 2yr. old races would be a good example. The same would apply to a lot of Stakes races as Mark has already shown. A common sense approach would be all that is needed to work out where and when the pricing method is useful or useless.

Your idea is handicapping the prices Bagman, although I [partially] agree and it could be taken a bit further for a mechanical approach there by separating the SR from a runner's maiden, C1 to 1MW and Handicap wins if you wanted to, but then we are really shifting into Handicapping for the purpose of selection.

I think the SR by itself is a good reflection of the horses overall ability and the trainers overall ability of selecting suitable races. If a horse is starting in the winning odds price range [say $2 - $6] that creates punter interest, the SR as a pricing guide [without breaking it down for several different classes of races], should still hold up well as a price indicator, by itself. A small break down as you suggested though, would be useful but common sense would be the best guide.

We should remember again that the method is to get an idea of SP value only and it is not meant to become part of the selection process of Handicapping, which is what a lot of readers are already or will automatically[subconsciously from well used habit] try to do and miss the point and the idea's value completely.

[crash]

I know what you mean Blocka, and of course you are right [as was Mark], but the idea is not to work at or arrive at a market price for the field or to create a book, but just to see the relationship of your selection[s] actual value as a bet.

If a selections SR is 30% we divide that into a 100 and we get $3.33. That figure represents it's overall true value. Common sense then works out a suitable price that would reflect overlay value. It could never be purely mechanical of course.

[Shaun]

Hey Crash i like that...sounds good to me....and like you said it is a way to find overlays you could esily use it just on the one runner you select and on some races it just won't work...but we don't have to bet on all races

[xptdriver]

Gday All

I have had a look at what Crash is saying here and it works pretty well... As some have said earlier it doesnt work on all races... but with a bit of thought It does work and you can apply it to all races where there are horses with a win % (obvioulsy).. For someone who does ratings, I think this could be useful if you say took your top 4 as you rated them and then compared them to Crash's idea.. and bet the overlay accordingly.

Before I get accused of backfitting.. these are my actual ratings for last in Sydney today

4 $4.9 DESERT DANE
7 $4.9 COUNT LUSKIN
5 $6.5 HOT COFFEE
11 $8.7 PENNY OPERA


Using Crash's idea the prices you need to back any of these horses is as follows..

7 1318 Count Luskin (2) $2.00 BET
4 04x5 Desert Dane (16) $2.38 BET
5 116x Hot Coffee (18) $4.76 SCR
11 78x9 Penny Opera (NZ) (7) $6.67 BET

For the record Count Luskin won paying upwards of 9 bucks.. and in that race it was the horse with the best SR clearly...

There were other races thru the day where it seemed to work well

Crash, this idea is one I never even considered, but it shows promise I reckon if you can apply filters... I think the top 4 in your ratings (or you could go wider or narrower for that matter) is probably not a bad place to start..

Thanks for the food for thought

[crash]

Interesting example XPT.

The SP prices reflect the 18 runner field size, so I wouild be looking at adding from 75% to 100% to the prices needed before calling any price an overlay and even if 100% was added there are still some among your runners there.

Working out a simple %+ guide for various common race sizes would be pretty simple. Even for a 6 horse race though, I would want some % over to represent profit on a runner's actual 'odds' shown after converting from SR.
At least we wouldn't be betting blind value using this method.