Overstaking...is it the biggest downfall of most punters?

Being one of the "Most Punters" group is their biggest downfall.

If you stay in that group then best thing is to enjoy it as a recreational pastime, you will have
winning streaks and losing streaks. Enjoy the good times.

If you want to get out of that group, It's a journey of hard work, set backs and learning.
A topic rarely mentioned on that journey is how important it is to learn more about yourself, your strengths and weaknesses.
The more you know about yourself, the more you can play to your strengths and rein in your weaknesses.

Becoming a winner is about realising your biggest opponent is yourself.

Agree with all that Escapee, temperament and discipline are paramount in all aspects of a successful life.

Morning G Hall sent you a message on Horse Racing Form at 10 am this morning or The top of the Page here Thanks

I’m not the racing blogger

biggest single bet I’ve ever put on is £500

an lost

emotionally never felt good with that amount on

I can handle £50 e/w max

biggest downfall of most billy bunters is that they bet the team/horse/player, not the price.

I dont think overstaking is the biggest problem but overtrading. I think nearly all losing gamblers are addicted to the action. If they can lose an acceptable amount then they will continue. They are basically addicts addicted to the action.  They know they will never win long term but they dont care as the action is more important.

I like the replies above.

Understaking is the problem.
Thinking favourites and thinking outsiders is another problem.
What others think indicates you do not have a sound self-generated opinion on every runner.
Betting often is a problem, as you would not have done deep analysis on all the bets.

I gave up betting in October 2021.
My practice was to increase the bet the higher the odds.
Of course I had to have a good reason to back the horse, a reason few considered.
Bet on the exchange, and win only.

Imagine that you had a bank of £100 and you were offered bets on a fair dice that paid out 6-1 if you picked the right number. You could play the game as long as you wanted, unless you lost your bank. The correct staking would be Kelly, as this would maximise the long term winnings and stop you going broke. On the other hand, if you were only allowed 10 bets under these conditions, you wouldn't be using kelly (I'm not sure what staking would be correct but it might even be right to put the whole bank on the 10th and last bet).

Most bettors are nearer the latter condition than the former and if they find a very good bet they probably should put as much as they can on it.

This might sound like a boast or the greatest aftertiming ever.
Fwiw I think I have an edge in horseracing that few have, or few consider, or few would do the work to gain the edge.

I analyse the pedigrees to six generations (occasionally ten generations) of thoroughbreds racing on the flat.
See my Wordpress website foalmare.com (not a commercial sire, I sell nothing).

Back in 1998 I was travelling from Dublin to the Oaks and Derby at Epsom.
These were the first two races where I analysed the pedigrees of the fields.
I bought a book from New Zealand the previous month (The Thoroughbred Breeders' Handbook by Clive Harper)
I backed Shahtoush at 25/1 in the Oaks (won) and City Honours 11/1 in the Derby (2nd btn a head).
The next time I analysed fields was two months later and I backed consecutive 16/1 winners in Group races.
This is not a person from China telling you he is good at picking winners, "please subscribe".

I believe pedigree analysis gives an edge in these conditions
(1) flat races, preferably at 8f and upwards for 3yo+, or 7f+ for 2yos.
(2) ground conditions must be at slowest good/soft (the analysis suggests speed)
(3) maidens/unraced; conditions races; listed, group races.

This is not an easy task.
Data collection is a real burden.
One thing it does is force you to back horses at high prices as you are doing what no one else does.

kincsem what do you look for in pedigree data?

I sometimes used to follow the 2&3yo foals of a star horse, but to go back 6 or even 10 generations??

I wouldn't know what I was looking for

I'll start with what I don't look for.

(1) forget black type on the dam line.
I have nothing against it but many of the horses in the catalogue (by the dam, by the 2nd dam, by the 3rd dam, and even by the 4th dam have nothing in common with the sale lot (except their dam).

I analysed the subsequent racing performance of all 2,109 yearlings sold at a UK yearling sale.
There was almost no relationship between money paid to buy them and money earned on the racecourse
e.g. 29 of the 30 most expensive earned 3.5% of what they cost, a 96.5% loss (the other one race in Hing Kong).
That buying performance should tell you about experts who stroke their chin and mutter things about ......
....... comes from a good family, the stallion stamps his stock, he walks well, stands over a lot of ground, nice sloping shoulder and other filler.

(2) forget sales averages i.e what was the average sales price of that sire's yearlings last sale / last year.
(3) forget catalogue updates e.g. a half brother won a listed race in France last week.  Who cares?
(4) I don't care if the first runner by a sire won first time out. That is just stage managed to fill newspapers.
(5) You read that stallion A's yearlings sold for 240,000; 200,000, 180,000 guineas. Sure, but who bought them? Did the stallion owner get his mates to buy them and he bought a few from his mates in exchange?

Of course if you want to sell to the next fool then get him to believe the above - the greater fool theory - you sell to the biggest fool.
You might make money selling dud horses to trusting buyers.


From my work I believe only about 25% of a runner's ability is inherited from its dam.
By that I mean the dam's black type is not inherited or her rating is not inherited (a very small amount is.)
A dam rated 110 will probably produce runners rated about 80, which is above the average 70, but not much above.
The real attraction of a dam rated 110 (or similar) is that buyers believe (incorrectly)her foals will also run to a rating of 110.
Why do they believe that?
Because they are assured that is the case, and there are no statistics to tell them otherwise.
Why would the breeding industry tell the truth when they can sell at high prices.
They are sellinh hope, not selling horses.

Forget the sire's racing ability.
Almost all of them won a Group 1 and had similar ability. Only about 1 in 200 sires persists in pedigrees beyond a few generations, the giants of the breed.

Now that I have dissed almost all that everyone believes what do I believe?

I've done all the usual stuff.
I even prepared a database of all IRE, GB, FR, GER, ITY, USA Group race winners from 1900 to 2015 (USA Grade 1 only).
About 11% of Group race winneres are out of a dam who won a Group (or Graded) race.

I ignored Listed winners as there are as many of those as Group 1,2,3 races combuined, and I ignored placed G1, G2, G3, Listed horses.
The sales companies love to "award" black type and doubling the number of races (if you include Listed races) and then trebling that number (if you include 2nds and 3rds) and you have thousands of animals with black type.
A runner that won a Listed race out of the 3rd dam might read well, but if you look closely that Listed winner was almost certainly by a sire that has zero genetic connection with the sales lot advertised.

What works?
Inbreeding works, but it works both ways, negatively and positively.
Unfortunately modern breeding practices (from 1970s to now) have greatly increased the negatives, making it much less likely you will breed a good horse (or find a good pedigree in a race).

You might need to pay attention here.
The best sire of the 20th century may have been the Canadian bred Northern Dancer (1961) (m)
His breeder offered him for sale at USD 25,000 but none of the experts bought him.
He was too small.
But Who would you bet on in a race between a hare and an elephant?

Whatever.
The owner used him as a stallion and eventually his fee rose to USA 1,000,000 a mare due to the tremendous success of his runners (fee rose from a couple of thousand).
In my data Northern Dancer (1961) (m) produced 522 offspring, 234 males, 288 females.
201 of the 234 males were bought by others and used as stallions, the new owners thinking their sire must inherit the ability of Northern Dancer (1961) (m) and be equally good as stallions.
A few were very good: Danzig; Sadler's Wells; Lyphard; Nijinsky; Nureyev; Storm Bird.

This is where the problems begin.
Danzig produced 742, and those 742 produced 18,599, and those 18,599 produced 50,021 (in my data)., and the other sire sons of Northern Dancer also produced tens of thousands within a few generations.

For Northern Dancer (1961) (m) the numbers were 522 (I call this prog1 for progeny in the 1st generation i.e. sons and daughters;
Those 522 (prog1) produced 19,683 (prog2) ....... and the 19,683 produced 103,831 (prog3).
These huge numbers happened because greed took over.
Previously stallion owners (pre the 1970s) stuck to a self-imposed limit of 40 mares a stallion a year.
Now it is common to send 200+ mares to a top sire and even in Ireland 300+
More covers equals more fees, and money is king.
When the Northern Hemisphere breeding season ends (15th February to the end of June) they fly the sires to Australia, New Zealand, South America and they repeart there what they did here.

Here is the punchline: a stallion can be the father of many in a year, a mare can only be the mother of one.
The 103,831 great-grandchildren Northern Dancer produced in three generations were 97,212 through his stallion sons and 6,619 through his daughters.
This is a ratio of 14.69 sons of Northern Dancer in the ancestors in pedigrees to every 1.00 daughter of Northern DAncer
You might say sons are stronger, are better runners, and that imbalance is a good thing.
You might also say I got it wrong and it must be 50% male, 50% female - think about it.

For a good male runner the best simple inbreeding group is a son and daughter of a duplicated sire.
By far the most common inbreeding group in the 21st cebntury is two or more (often many more) sons of s duplicated sire, usually Northern Dancer or Mr Prospector.
What you get with sons of males in the ancestors (4th, 5th, 6th generations) is a slow horse.

What do you need for a good filly runner?
You need two or more daughters of a duplicated sire.
With one daughter of Northern Dancer a rarity in pedigrees what do you think are the chances of getting two daughters of Northern Dancer (or three)?

For simplicity I used Northern dancer as an example.
Of course you can have dams duplicated in pedigrees, but as females produce fewer offspring that is not as common.

Northern Dancer (201 stallion sons); Mr Prospector (266 stallion sons); Danehill (159 stallion sons); Sadler's Wells (166); Danzig (174); Seattle Slew (121).
This is a rush to profit from the reputation of great sires by using their sons (cheaper to buy but not as good) as cash generating machines.

The best inbreeding in pedigrees if full siblings in the ancestors in the 4th, 5th, 6th generations.
The best is a full sibling sister and her full sibling brother, and better still if she produced a son (opposite sex to herself) and the brother produced a daughter (opposite sex to himself.)

Inbreeding in the 2nd and 3rd generationm ust be avoided. It is too close.

Best for a colt runner
(1) son and daughter of a duplicated sire
(2) two sons of a duplicated dam
(3) extra of (1) and/or (2) or both
(4) full siblings or 7/8 or 3/4 siblings (not 1/2 sibings)

Best for a filly runner
(5) two daughters of a duplicated sire
(6) son and daughter of a duplicated dam
(7) extra of (1) and/or (2) or both
(8) full siblings or 7/8 or 3/4 siblings (not 1/2 sibings)

But what is in almost every pedigree on the racecourse now is two, three, four or more sons of a duplicated sire(s).
I call this "a recipe for slow".

It sound simple.
But pedigrees now are so clogged with sons of a small number of sires that you would need to sift through tens of thousands of stallions and mares to get a combination (foal) with the above.
What I try to do is plan a pedigree with both colt runner and filly runner "factors" i.e. (1)(2)(3)(4)(5)(6)(7)(8)
Of course (4 )and (8) are the same i.e full siblings.

Full siblings in the ancestors is the magic.
Try to avoid two full sibling brothers producing sons - males producing males - not a disaster but there is better.


..............................................................................................


Your question
kincsem what do you look for in pedigree data?

You will find sfa data in newspapers, the internet, or even in breeding books that is any use.
Industry stats are useless, just telling you who won what, and giving meaningless percentages.
You want to know the reasons.

Why is one brown horse faster tha another?
People claim they can tell ability by looking at the bodywork (conformation).
You need to look under the hood at the engine (the pedigree).
Of course pedigree is about probablilty.
Two horses with the same pedigree might be entirely different animals - if you are a male and have a few brothers did you expect all brothers to be exactly the same?
No guarantees, but if you are fishing for fish you want to be fishing in waters that contain fish, and you might still catch nothing.

My best breeding books (I have about 1,000 books on pedigree and racing) are 40 pages, 56 pages, and 110 pages.
The other books waffle on about this race he won and his mother won another race and you go around in circles learning nothing.

If you want an edge you will have to roll up the sleeves and work.
Read my Wordpress website foalmare.com
If it is too difficult to understand or you want solutions handed to you you stay where you are.

I wrote a long reply of about 40 lines and when I hit "post reply" it disappeared.
Perhaps the mods want to review it or binned it.

You need a six generation pedigree of a horse and look at the duplicated ancestors in it (the duplication groups).
You would need to go to pedigreequery.com and print out a 5-generation pedigree of the sire, then print out a 5-generwtion pedigree of the dam.
place the sire pedigree on top and the dam pedigree below.
that is a six generation pedigree of the horse.
Use a colour highlighter to colour the duplicated horses (a fe different colours for the different duplication geoups).

If you want an example look at the 5-generation pedigree of my filly foal Chena (2025) on pedigreequery.com and look at the coloured duplication groups.

Or look at Chena's 6-generation pedigree here
https://forum.thoroughbredvillage.com.au/interesting-pedigrees_topic69442_page1.html
look at this post ....... Posted: 22 May 2025 at 1:38pm
Below the post you will see a listing of the inbreeding duplication groups

Chena (2025) (f) died from colic at five month old.

I know the thread is about overstaking.
I hijacked it to talk about something else - an edge.
You are overstaking if you are betting on what everyone knows, what you read in the racing papers and on racing websites.
The value is gone, everyone knows what you know, and the price is lower reflecting the many punters who followed the inforrmation.

The margin in flat racing is / or was about 17% negative.
Bet as the crowd bets and 100 becomes 85, 73, 62, 53, 46, 39, 33, 28, 24, 21.
My opinion is you are overstaking if you are using information others use.
If your information is easy to access it is of no use.

I thank you for your extensive reply kincsem, unfortunately (for me) I glaze over, I've tried reading it twice, but
whilst I know there is some sense and insight... it escapes me.

If you can download £money from the internet based on reading 3, 6, or 10 generations of genetic form I tip my hat in
admiration and respect.

With my recreational punter hat on, I love the buzz of finding a horse that sold for £100k-£200k as a yearling and sold for £10-20k as a 2yo running in a class 7 novice on all weather on a Tuesday afternoon at 20/1 or even 50/1.

My theory is that 'pedigree' info is in the price paid as a yearling, and that ricks happen when the stable that bought it sell it on cheaply as a 2yo.


With my pro punter hat on.... it's all about the rule of large numbers, I was curious if you had any "large numbers" (i.e. 1,000's of races) insight into multi generational "pedigree"

Escapee
With my pro punter hat on.... it's all about the rule of large numbers, I was curious if you had any "large numbers" (i.e. 1,000's of races) insight into multi generational "pedigree"

The insight is on foalmare.com and the data sample was 159,200 rated horses compared to the inbreeding duplications in the six-generation pedigrees of each horse.
See (table 6) (table 7) and (chart 3) (chart 4) and especiallt (chart 5).
The link between the counts of different inbreeding and ratings was verified using the Chi Squared stat function
i.e. the link is real at 95% and/or 99% confidence (or it is a 20/1 or more chance the link is false).
The ratings I use are the 3yo and older year-end rating from about a dozen different sources in IRE, GB, FR, GER, USA
Often I have three or four rating for one horse so I use the highest.
The ratings are not compatible, but all I want to know is the general level of a horse.

You will NOT find anywhere a study that analysed 159,200 pedigrees against ratings.
The NZ writer I admire greatly analysed 1,400 yearling pedigrees in the early 1950s (he analysed three annual yearling sales before the sales).
He wrote out pedigrees by hand.
I did that in 1993 and it took 15 minutes to write out a four-generation pedigree.
Writing out a six-generation pedigree (126 ancestors) would take an hour.

A program I wrote about five years ago generates 195 ten-generation pedigrees of 2,046 ancestors (16 times larger than six-gen) each second.
Over the last month I worked on a faster 10-gen pedigree generation program.
It might be twice as fast, and I also had a new fast PC built last month (collect next week) for more speed.


The lack of a strong connection between the running rating of a dam and the running rating of her offspring had a sample of over 65,000 (see (table 3)
(i.e. the dam and her offspring both had to have a rating)
Many dams were unraced, or many of their offspring were unraced.



Why would a buyer pay big money for a yearling?

If approached by a new owner the agent will find out how much the new owner is willing to spend.
He will spend that to the limit because if your commission is 5% of the price you have an incentive to pay the maximum.
Even worse is a budget of 50k might be spent on a horse worth 10k.
The buyer could arrange for a few people to bid up the horse to 48k and then he will step in and bid 50k and "secure" the horse against stiff bidding opposition.
There could have been an arrangement to split the "extra" 40k 50/50 between the breeder and the agent, i.e. an extra 20k each.
The breeder gets the 10k that the horse was worth, and the agent gets 20k and a commission on 50k.

Of course that never happens.

* The breeder gets the 10k that the horse was worth and an extra 20k, and the agent gets 20k and a commission on 50k.

Scepticism is the attitude of doubting knowledge claims, questioning evidence, or suspending judgment due to the inadequacy of evidence, often requiring rigorous proof.

You need to do your own analysis.
If you take opinions offered by newspapers or websites you are taking low odds because those opinions are public.
Areas for work are the draw, going, racecourse characteristics, field sizes, pedigree, and probably half a dozen other factors.

Draw - forget trying to figure out if high is best or low is best. Forget the draw on straight courses.
Did you know that stall 10 in the English Derby won about eight times from the introduction of stalls in 1967 up to 2010, and stalls 11 and 12 never housed a winner?
Did you know that of good ground or faster low draws 2 to 7 in The Prix de l'Arc de Triomphe are highly favoured, the faster the ground the bigger the advantage.

The reported going is almost always incorrect, and is the first thing you should bin. Race times tell the real going. You might be betting on a horse unsuited to the going
Almost without exception japanes horses in the Prix de l'Arc de Triomphe have spent their entire careers racing on firm ground on pancake flat tracks - no surprise they bomb in the Arc.

Is the course uphill/downhill, galloping, round tight bends, undulating - not knowing what suits your pick and you are overstaking.

Have you ever considered the wind?
Look at the ride Frankie Dettori gave Blue Bunting in the 2011 1000 Guineas.
Drawn 16 of 18 i.e. on the very right of the field as you look towards the finish Dettori took her back and to the very left of the field out of the wind coming strongly from the right, winning on the stand rail.

People doing complex calculations using weight differential, beaten lengths, speed ratings, collateral form are ignoring plenty, probably more important than previous form.