Winning with Patternform, a logical approach

Let us know how you use Patterform to help newcomers
martinkil
Posts: 3413
Joined: Sun Jan 11, 2015 9:55 am

Winning with Patternform, a logical approach

Postby martinkil » Thu Jan 15, 2015 8:40 am

A very old post which may still have relevance today


Joined: 30 Oct 2005
Posts: 6692

Posted: Fri Oct 27, 2006 5:22 pm Post subject: Winning with Patternform, a logical approach

--------------------------------------------------------------------------------

This article was first posted by Aubrey in the last Patternform forum. I hope he doesn't mind this repost:


I thought it might be helpful to a few folk to describe in broad terms the method I use to pick horses with patternform. Patternform is unique in that you can adjust the sample of races you compare to match the race in question. Whereas ordinary ratings are averaged from a series of races for each horse that might not really be sufficiently
similar for an average to be valid at all.

My method can be described as a series of step by step processes but before everything else the first thing to understand is the priority of importance in selecting winners using any kind of form study.

Here are Brian Goughs excellent thoughts on the matter:

The horse must be fit and well.

1.The horse must be in form. (careful with this).
2.The state of the prevailing going must suit the horse.
3.The type of race track must be suitable.
4.Is the horse still improving ( be aware of the w.f.a. improvement).
5The distance must be right, although some horses might have a broader band of suitable distances than others.
6.The horse must be running against horses of similar or lower class than in it’s last race. This might exclude horses that won it’s last race and are progressing up the class ladder as a winner.
7.The pace of the race must be genuine. This is something you will have to estimate based on the class, distance, number of runners and if there are known pacemakers.
8.There are other factors e.g. the draw bias at the track, if relevant, and the jockey and trainer form. Whether the horse has the ''right'' jockey on board. Also sometimes the age of the horse. Watch out for first time blinkers. Perhaps trainers travelling a long way to run their horse.\r\n\r\n\r\n

SO considering the above, use patternform to determine which horses fit the bill.

Is the horse in form?
You can check a horses most recent races if possible within a month but in any case as recent as possible to see if it is running close to its best form. If a horse has NOT run for some time then it depends on the trainer as to whether the horse is likely to be in race form first time out. Some trainers send horses out first time in a season "for a run" while others usually those with better facilities send horses out fully race ready. Check the "Longest travellers" on the racing post site no trainer is going to send a horse 100 miles just "for a run" unless perhaps as a companion to a star horse from the same stable. If your horse has not run for some time but has travelled alone or with a known low grade horse more than 100 miles you can be pretty sure the horse is coming to the "office" for serious work

Check the going right this minute!
Is it as expected the night before? you can check going by listening to the commentary or reading the official going on the result of the previous race remember that the official going is always a bit behind actual going. if its raining then the going will be drifting softer if its sunny and especially if sunny and windy the going will be drifting to firmer. This is especially useful for LAYING horses select going filters that lean toward the expected trend of the drift in going when looking for WINNERS and select filters OPPOSITE to the trend in going to find SHORT ODDS LOSERS for laying. NOTE!!! if the top of the list horse Does not change when the going filters are swapped around this way, the horse is an all round performer and is likely to win not only because it can handle going but because a horse that can do this is what the industry calls "genuine" i.e. a horse that always tries their best.

Check that the horse has done well on a track of similar type to the track of the race in question click the form line and look at the horses best and worst performances on the different tracks and then compare that to the track types.
Only registered users can see links on this forum!
Register or Login on forum!

%20

If the horse or its close rivals are less than 5 for long races and less than 4 above the next best horse for shorter races watch out for improvement due to the weight for age scale.
Only registered users can see links on this forum!
Register or Login on forum!



In theory Martins weight adjusted figure should compensate for weight improvement but it is a risk factor worth using to exclude riskier bets.
Check that the horse has either run the race distance well or is likely to. If it has not run the distance then to find out if it is likely to check that it was either in front in the furlong before the post if the race in question is shorter than before or that it was making up distance and running on strongly if the distance is longer than before. Check the date and place of the race you want to check your horse out for and go to a form site such as sporting life click the racecard for the race then the horse then the race you selected in the horse race list to get the notes for that race.

5, 6, 7, and 8 in the notes above are self explanatory and are worth checking out too if you have the time or are serious about making a profit rather than just keeping your betting expense down. One last thing is to watch the market especially close to race time. If a stable has a horse they think is likely to run better than its rating would suggest the chances are they will bet as late as possible and then in big money so watch for a sudden rush of money for a horse you are not planning to back. if that happens either don’t bet on the race or at the very least place a cover bet on the shortening horse.

This as you can see is more complicated than a "system" but it is also much more reliable it just takes a bit more work. No system can deliver the sort of results you can get if you properly investigate the runners in a race and choose the races you bet on based on how much you know.
The shortcuts of using a rating system "blind" will help to find the odd winner and probably enough to make betting marginally profitable long term but doing your own race analysis based on the rules above will be much more profitable and more importantly far less prone to "losing streaks".

Here is the step by step process for FLAT horses.
(Jump races are easier but I''ll cover that in another post if folk are interested in this one)\

1- with all filters open check the top rated horse at the full life of the oldest horse and for 24 and 12 months in the race. This gives a baseline for you to consider all the filtered values against.
2- Next check the top rated horse in the last month. This tells you which horses have run within the last month and whether their rating was anywhere near their best.
This first step gives you a basic picture of the horses in the race. One thing that you should note right away is that if one or two horses appear consistently in the top of the list these are likely to feature near the front of the current race.

3-Next check the going filters initially in the following way. IF the going is listed as good to good to firm then select gd, gf,fm. if listed as good to soft then select gd g-s sft hvy In other words select all going tending in the direction of the prevailing going. If the going is good to firm or firm also select std as horses that go well on AW are better at G-F than other goings.

4-Repeat the previous exercise 3 for different months back as in 1.
Watch the top three horses as you do this. if one or two are always there then your beginning to narrow in on a selection. If the top horse is the same at 12 and 1 month then:-

5- Click on its form line at 12 months. Look at its performances and check for consistency is it fast one time slow the next? if so are the slow times on a particular type of track? are the good times on a particular type of track? Check the tracks against the reference above. Sometimes the difference is right and left hand track preference sometimes tight or galloping. Sometimes the horse is just plain inconsistent. If so and the horse is capable of fast speeds the trouble is this horse can not be ignored as a threat to another more consistent horse but it is not a good bet in which case it may be time to move on to evaluate a different race.

6-Check the going for all the races the horse did well in and make sure it will act well on the going in the current race. NEVER underestimate the importance of going it has opposite effects on different horses and so is the single biggest reason for unexpected results in racing. Even just the difference of a sunny afternoon firming up a course can change the expected winner and as a patternform user you will be one of the few who know which horses are capable of taking advantage of the prevailing going and to what extent so can often expect a good price win.

7-HORSE CHARACTER AND WINNING HEART.
The greater the difference between the top rate horses and the next top in their most recent performances especially the greater the basic chance of a win. As the differences get closer though other factors become more important. For example does the horse have the heart of a winner? In a maiden of course none of the horses will have won but you can check your top two horses race notes on sporting life to see if either made an effort to win. For horses that have won did they beat a horse in a fighting finish? if so good that’s more important than winning distance.

Ideally there will be two horses in the race that are substantially better than the rest and one horse a bit better or with better winning skill than the other. This oddly is better than a race where there are a lot of slow horses and one fast horse. Horses have a strong herd instinct and do not like to be out on their own. If the horse has a jockey that is a good enough rider for the horse to consider him/her an "honorary horse" (No I’m NOT joking) then this is not such a problem but if the jockey does not have that skill then the horse will tend to lose concentration as the sound of hoofbeats fade into the distance. You will notice horses are happier out in front on their own on quiet days than windy days.
I used to find that horses that knew me would quite happily fly away from the pack but a horse I had just met that morning would suddenly get uncertain and come on and off the bridle if allowed too far out in front. The snag is if the horse has no finishing speed you HAVE to go out in front and so its handy to have at least one horse prepared to come with you. This means that a sequence of horses at the top 70, 67, 58 is better in the sense that the horses arrive in expected order than a sequence 70 60 58. Strange but you will find it to be true.

8-IMPROVING THE RACE SAMPLE
if the race in question is a medium to high price race (20,000 to 70,000) then exclude prize money of less than about 7000 from the sample. This might knock off a winning horse but it more likely to knock off horses rated inaccurately in low grade races that would tempt a bet and then flop in the race. You can also use the class filter to exclude all but one class down and all classes above the race in question. A horse moving up one class has a chance but a horse moving up two classes has virtually none even if its rating suggests otherwise. Horses moving down class have a big advantage. (imagine if you were racing other people every week and for weeks and weeks you had no trouble keeping up or getting to the front and suddenly one day you race and find that you are puffing to stay with the field? you get demoralized and might give up. horses feel the same! In the same way if you have been puffing and panting to keep up for weeks and suddenly you are in a race where you are strolling to the front effortlessly it gives your feet wings!! Again horses feel the same!

9-If the race is a HIGH priced race or a classic then pay far more attention to the long term statistics. In high priced races high quality horses that rarely race are more likely to be involved and can be expected to present for a race in top form even if they have not raced for some time. In lower grade races what the horses did last week is more important than what they did a year ago but in a top grade race that is not so.

10-Clearly if the rating column is filled with the first ten horses less than 10 points apart you just don’t bet on that race. Deciding how much of a margin makes it worth a bet is harder though. In the case of younger three year olds watch for improvement so two horses that are the same rating but from races a month apart will tend to favour the horse that raced earlier since it can be expected to improve according to the weight for age scale. In to get a feel for the real meaning of the rating remember it is basically notionally in pounds.

This is a ROUGH guide again taken from Brian Gough''s excellent site(you will find it on the links page)\

5 to 7 furlongs 3 Lbs. per length
8 to 11 " 2 Lbs. per length
12 to 16 " 1.5 Lbs.per length
16 + " 1.0 Lb. per length

This guide gives you an idea of how the ratings relate to actual winning positions in as much as they ever do! to start thinking a horse has a real winning chance bigger than the issues of heart or winning skill mentioned above you want the horse to be at least three lengths up. A top jockey like Mcoy can probably improve a horse by as much as five lengths compared to the run of the mill but more concern should be paid to real tyro jockeys they can slow a horse by ten lengths! Watch out for amateur riders. They ride longer stirrup usually and often have a much closer relationship to the horse. While often amateurs are not as good the best of them are and their closer relationship to the horse can be worth as much as a stone. That’s why amateur races string out more than professional. the difference in skill is one factor but also the close relationship means the horses don’t worry about being away from the herd. Horse rating differences in amateur races can be three or four times greater in lengths than in pro races. However these factors are more commonly going to occur in jump racing and so enough said here.

If you take all of these points into account you will rarely have a losing day. In the last month starting with a 60 pound bank and betting mainly 2 pound bets I have made 623 pounds using the method as described and that includes losses in races where I was just sloppy and forgot to check the obvious. I think with full care an attention I could have done even better.

I hope these notes will be of help to the other users of patternform and to beginners. Try to follow the order of the above as if you do you will avoid accidentally assuming that the differences in rating at the top with the filters ON is a true reflection of the horses with chances in the race. Watch that first screen as it tells you how close the horses are in general quality. If they are all really close skip that race for now and come back to it only if you are prepared to take the time to winkle out the winner by going back and checking race notes etc. Basic rating info wont be enough.

Scotty (Aubrey on the AM forum)

Back to top


Aubrey



Joined: 26 Oct 2006
Posts: 74
Location: Edinburgh
Posted: Fri Oct 27, 2006 10:21 pm Post subject:

--------------------------------------------------------------------------------

Hi!

Thanks for posting it again.
it is now easier to use with your added features So I might re write it with examples on use of class and pace info if folk are interested. The fact is that patternform is an instrument that is best used as an aid to a fundamental logical approach rather than as a straight rating system rigidly used as "bet on the top rate or top two rate" method. Used properly, as a tool to asnwer the questions relevant to a given race it can build you a really deep insight into the race and the variables likely to factor in to the result. It also so often shows you just how sensitive a race is to variables and so lets you avoid bets on races that are effective lotteries and also and even more usefully often highlights cases where you have a virtual certainty that a specific three horses will be the the first three home and can therfore bet in a place forecast and seriously clean up!

I love patternform!!

Aubrey

Back to top


efc1878



Joined: 21 Feb 2006
Posts: 8

Posted: Tue Apr 10, 2007 6:16 pm Post subject:

--------------------------------------------------------------------------------

Aubrey
i would be most interested in class and pace examples been using patternform on and off for a while now still workin it out i need to use it more often to make the most of it and start keepin some records.

Back to top


giorgio2



Joined: 05 Feb 2006
Posts: 34

Posted: Wed Oct 24, 2007 3:48 pm Post subject:

--------------------------------------------------------------------------------

not sure if aubrey is still around the forum, just been reading through is logical approach method, very interesting, in it he says he will produce a step by step approach for the jumps if there was any interest, any chance aubrey, cheers george

Back to top


sun69uk



Joined: 20 Sep 2006
Posts: 26

Posted: Wed Mar 05, 2008 7:19 am Post subject:

--------------------------------------------------------------------------------

umps here as well interested

Back to top


onefurlongout



Joined: 20 Oct 2008
Posts: 27

Posted: Wed Dec 30, 2009 6:47 am Post subject:

--------------------------------------------------------------------------------

Hi (again)

Could anyone shed anylight on the below comment?

1- with all filters open check the top rated horse at the full life of the oldest horse and for 24 and 12 months in the race. This gives a baseline for you to consider all the filtered values against.

I'm not sure what the full life of the oldest horse means.

Sorry for being a bit thick

Back to top


Martinkil
Site Admin


Joined: 30 Oct 2005
Posts: 6692

Posted: Wed Dec 30, 2009 10:41 am Post subject:

--------------------------------------------------------------------------------

I think it just means 24+ months, since this is the only time period available over 24 months

Back to top


mamoran



Joined: 07 Feb 2010
Posts: 2

Posted: Mon Feb 08, 2010 3:55 pm Post subject: Question on ratings

--------------------------------------------------------------------------------

Aubrey wrote:
Hi!

Watch that first screen as it tells you how close the horses are in general quality. If they are all really close skip that race for now and come back to it only if you are prepared to take the time to winkle out the winner by going back and checking race notes etc. Basic rating info wont be enough.

Aubrey


Hi , Im new here and have really enjoyed reading the articles. I have a question on the section above.
When I open the racecard for a particular race how can i assess a horses rating for the race. Is there a numeric value given for each horse and if so where is it, otherwise how can I assess whether the horses are close together in quality?
Thanks in advance
Mike

Back to top


Martinkil
Site Admin


Joined: 30 Oct 2005
Posts: 6692

Posted: Tue Feb 09, 2010 8:46 am Post subject:

--------------------------------------------------------------------------------

Hi Mike,

There are three main columns you can use for ability, they are Class, Speed, and RanTo which is a form rating. For each runner you're given it's top three rating from the races defined by the "look back" field. (default is to look back at races in the last 6 months)

So lets take the MARKET RASEN 3:30 (http://www.patternform.co.uk/marp095g.htm) race

On the default settings when you load the page the "look back" field is set to "6 months". On the Speed ratings (default sort field on loading) Carrickmines is top rated with speed figures of 57, 56 and 51.
If you click on Carrickmines form figures "f3112- 32" a gold box should appear and you can see that the 57, 56 and 51 speed figures come from the races on 28Jan2010, 03Dec2009 and 25Nov2009.

If you click on the Ranto title the runners will be sorted by RanTo (form) field and Furze Hill is now top rated with ratings of 52 43 42 with Carrickmines 2nd top rated with 51 51 49.

If you change the "Look Back" field to 24 Months and click Re-Run, then Hermano Cordobes is top rated on Ranto ratings with 74 72 70 but if you click on the form figures you'll see that those ratings come from 2008.


But to answer you question you need to look at the first speed ratings for top few rated runners and if they are close (in this case on default settings the top 5 have best speed figures of 57 56 56 53 amd 52) then the race could be competitive (use Ranto if you prefer form to speed figures)

Hope this helps.

Back to top


mamoran



Joined: 07 Feb 2010
Posts: 2

Posted: Wed Feb 10, 2010 11:10 am Post subject:

--------------------------------------------------------------------------------

You are a gentleman, thnks for the quick reply.

Back to top


Cowan6891



Joined: 11 Sep 2010
Posts: 30

Posted: Sun Sep 12, 2010 4:59 am Post subject:

--------------------------------------------------------------------------------

Hi

Could anyone help me with point 4 please;

Quote:
3-Next check the going filters initially in the following way. IF the going is listed as good to good to firm then select gd, gf,fm. if listed as good to soft then select gd g-s sft hvy In other words select all going tending in the direction of the prevailing going. If the going is good to firm or firm also select std as horses that go well on AW are better at G-F than other goings.

4-Repeat the previous exercise 3 for different months back as in 1.
Watch the top three horses as you do this. if one or two are always there then your beginning to narrow in on a selection. If the top horse is the same at 12 and 1 month then:-


I understand point 3 - I select the relevant goings.

In point 4 I believe I then go on to look at Full Life, 24 months, 12 months and note which horses are coming top.

But this sentance throws me Quote:
If the top horse is the same at 12 and 1 month then:-


Is the 12month 1month thing referring to FILTERS OPEN or with RELEVANT GOINGS?

Furthermore am I correct that if the horse is not top at 12month and 1month then the race should be binned?

Thanks
Ian

Back to top

barbaron
Posts: 5
Joined: Fri Jan 16, 2015 9:48 pm

Re: Winning with Patternform, a logical approach

Postby barbaron » Fri Feb 13, 2015 6:45 am

Hello mates,
Can anybody help me to understand correct the way to use this method, please? Please tell me what the author means under "rating"- it's Speed, Multiple, OR, or Ranto rating?
For example:
"2- Next check the top rated horse in the last month. This tells you which horses have run within the last month and whether their rating was anywhere near their best. "
Thank You in advance for the help, mates and wish all the best of luck!
Regards, barbaron

martinkil
Posts: 3413
Joined: Sun Jan 11, 2015 9:55 am

Re: Winning with Patternform, a logical approach

Postby martinkil » Fri Feb 13, 2015 8:15 am

The ranTo rating is the "form" rating. By that I mean that it is derived by looking at how far a runner finished behind the winner, the weights carried, etc...

The speed rating is derived by the time a race took to be run in conjunction with track variants .. etc...
Hence a horse can win a race yet return a poor speed figure if the race was slowly run

The Multi is a hybrid figure (derived by combining both RanTo and Speed) which a user suggested back in the dim and distant past

Hope this helps

barbaron
Posts: 5
Joined: Fri Jan 16, 2015 9:48 pm

Re: Winning with Patternform, a logical approach

Postby barbaron » Fri Feb 13, 2015 3:49 pm

Hello Martin,

Thank You for the fast reply of my initial question! I have read the help page, where are described Speed, Ranto, OR and Multi ratings, but still I'm not sure the user Audrie about which one rating talk in his post- [i]is this Ranto rating/because Audrie says 1.The horse must be in form.[/i]/? Any idea, please? Also I am still new in researching /paper trading/, I think this one method is good to start in horse racing analyse. Can You point me where to start, Martin?

Appreciate Your and other help here with this quality forum!

Regards, barbaron

martinkil
Posts: 3413
Joined: Sun Jan 11, 2015 9:55 am

Re: Winning with Patternform, a logical approach

Postby martinkil » Fri Feb 13, 2015 4:20 pm

Aubry writes:

Is the horse in form?
You can check a horses most recent races if possible within a month but in any case as recent as possible to see if it is running close to its best form. If a horse has NOT run for some time then it depends on the trainer as to whether the horse is likely to be in race form first time out. Some trainers send horses out first time in a season "for a run" while others usually those with better facilities send horses out fully race ready. Check the "Longest travellers" on the racing post site no trainer is going to send a horse 100 miles just "for a run" unless perhaps as a companion to a star horse from the same stable. If your horse has not run for some time but has travelled alone or with a known low grade horse more than 100 miles you can be pretty sure the horse is coming to the "office" for serious work

In theory the see if it is running close to its best form could refer to either speed or RanTo although due to the nature of the ratings the RanTo rating is usually the most consistent

barbaron
Posts: 5
Joined: Fri Jan 16, 2015 9:48 pm

Re: Winning with Patternform, a logical approach

Postby barbaron » Fri Feb 13, 2015 7:51 pm

Thank You very much for the patience and helpful explanation, Martin!

Regards, barbaron

weee86
Posts: 24
Joined: Sat Jan 17, 2015 2:22 am

Re: Winning with Patternform, a logical approach

Postby weee86 » Mon Apr 13, 2015 3:21 pm

martinkil wrote:Aubry writes:

Is the horse in form?
You can check a horses most recent races if possible within a month but in any case as recent as possible to see if it is running close to its best form. If a horse has NOT run for some time then it depends on the trainer as to whether the horse is likely to be in race form first time out. Some trainers send horses out first time in a season "for a run" while others usually those with better facilities send horses out fully race ready. Check the "Longest travellers" on the racing post site no trainer is going to send a horse 100 miles just "for a run" unless perhaps as a companion to a star horse from the same stable. If your horse has not run for some time but has travelled alone or with a known low grade horse more than 100 miles you can be pretty sure the horse is coming to the "office" for serious work

In theory the see if it is running close to its best form could refer to either speed or RanTo although due to the nature of the ratings the RanTo rating is usually the most consistent



This makes for a very interesting point.When you say close to its best form,in terms of speed/ran to, how many points of distance to the best speed/ran to would be a good mark?Do you think looking at course/going/distance should be important when making that difference?

weee86
Posts: 24
Joined: Sat Jan 17, 2015 2:22 am

Re: Winning with Patternform, a logical approach

Postby weee86 » Wed Apr 22, 2015 2:17 pm

Regarding last post,would 5-10 points from best ran to a decent mark?In the part where it says " If a horse has NOT run for some time then it depends on the trainer", how long do you consider " some time"?



Cheers,
Carla

martinkil
Posts: 3413
Joined: Sun Jan 11, 2015 9:55 am

Re: Winning with Patternform, a logical approach

Postby martinkil » Wed Apr 22, 2015 3:04 pm

For the "horse has NOT run for some time " I would use the -
Look back filter setb to - 24+ months
and the "Rest days" filter to whatever time it's been off - so for new flat season I would have 85+ days (85-183, 184-365, FTO/Unk)

Then you can see how the horse has run after a break before

weee86
Posts: 24
Joined: Sat Jan 17, 2015 2:22 am

Re: Winning with Patternform, a logical approach

Postby weee86 » Wed Apr 29, 2015 10:35 am

When a horse didn't win or place on a certain going, how can one know he ran well,by looking at,for example,the ranto rating?Is there some kind of minimum rating that tells that a horse ran a decent race ?


Return to “How To Use Patternform”

Who is online

Users browsing this forum: No registered users and 7 guests