Average Prizemoney

[Luckyboy]

G'day All,

I was wondering if I could get a few alternative opinions regarding Average Prize Money per race and using it as a selection filter.

I currently use Average Prizemoney as a 'culling' filter to reduce field sizes, but my recent results are drawing me to question its validity.

It seems to come unstuck when a race contains several highly rated (Zipform) up and coming horses who have low Average Prizemoney per start.

Any thoughts from anyone?

LB

[Mr. Logic]

I take no notice of average prizemoney. Don't even check it out. You have mentioned one of the drawbacks. There are others including huge prizes for various restricted entry races.
It also misses out on form - a horse finishing sixth, beaten only a length and getting no prizemoney in a million dollar race will not appear so flash compared to the winner next start.

[Luckyboy]

Mr Logic,

Thanks for your reply. I know that Average PrizeMoney per start should not be considered in isolation of other stats, but I had not thought about omitting it altogether.

Are there any others out there that don't use this race stat?


LB

[Chrome Prince]

Luckyboy,

My research shows that the use of any ONE particular filter is usually not particularly beneficial.

If you combine "Average Prizemoney" with "last start distance beaten" and "drop or rise in prizemoney for this race compared with last race", then you have a fairly effective guide.

In my opinion :wink:

The importance of each individual factor or filter in relation to a horse's winning chances is the age old puzzler! :eek:

<font size=-1>[ This Message was edited by: Chrome Prince on 2003-03-31 11:49 ]</font>

[Luckyboy]

Chrome Prince,

Thanks for your insight. I have utilised AVP as a selection filter along with the Zipform rating and last two start placings.

I like your idea of distance beaten last start. With regard to the prizemoney rise or drop would it suffice (in your opinion) to use the 'class rating' suplied in Sportsman as an alternative?


LB

<font size=-1>[ This Message was edited by: Luckyboy on 2003-03-31 11:56 ]</font>

Luckyboy wrote:

> I was wondering if I could get a few
> alternative opinions regarding Average
> Prize Money per race and using it as a
> selection filter.

If you take note of Malclom Knowles, he will say that Prizemoney should never be ignored and ranked average prizemoney is superior to ranked total prizemoney.

He published a booklet that describes the relationship between prizemoney and other factors, i.e., last start finishig position, distance, field size, class etc.

The booklet may still be avilable through Wiz Books, who were last know to be located in the ACT.

[Luckyboy]

Pasmaroo,

Thanks for your advice. I will look up the bookshop details in my latest edition of the Wizard.


LB

[noel]

luckyboy,
one of the problems with average prizemoney is that it includes bonuses which certain qualified horses earn, thus distorting the actual prizemoney earned purely from the race....cheers, noel

[Chrome Prince]

Luckyboy,

By all means use the class rating in the Sportsman.

The point is that AVP is useful when filtered and combined with OTHER factors.

I have found that rise in race prizemoney includes many winners at longer prices, while a significant drop in race prizemoney has a much higher strike rate with smaller average win dividend. It just depends on the approach you wish to take.

Another little factor is where the bulk of the prizemoney came from...was it a one off win two preparations ago, or was it a consistent build up each preparation. This can also be a huge difference in the end result.

Good Luck.

[Luckyboy]

Chrome Prince,

Many thanks for your advice. I spent a little bit of time last night re-looking at my filters and found approximately 52% of winners come from the top 6 ranked horses according to AVP (measured Saturday only last 18 months).

Therefore, I would suggest that AVP is probably a secondary filter. Your thoughts?


LB

[Chrome Prince]

Luckyboy,

Yes I agree here.

Here are some observations I have made during my incomplete analysis to date which may help....

1. 3YO horses in general are bad investment propositions. Many are overrated and often go around at well under their true price. Much of the hype is regarding previous 2YO form which doesn't carry on as they get older....perhaps burnout???
Examples - Lovely Jubly, 0 wins this campaign and early on when resuming was very short prices.
Victory Vein, 0 wins since resuming and first up was around $2.80, since then has gone around at $7.00 and $6.90 - going on this preparation form should have been at least a couple of points better.

2.Many punters look to improve strike rate or only consider strike rate of certain factors, but the most important part is setting minimum and maximum price filters as a higher strike rate may not necessarily relate to profit.

3. Getting the right balance of factors is the secret and key, but you'll never have a perfect record - it's basically impossible.
The only way to end up on the positive side of the ledger is to set min. and max. prices along with a reasonable strike rate and profit on turnover.

4. The amount of past data worked with has to be substantial. I have been guilty many years ago of pouring over a month of Sportsman's trying to find a pattern or system. This will not work and is a waste of time unless you have access to at least one year of data (preferrably a lot more).
It's like trying to run a census on the city of Hobart and applying the results to the whole of Australia!

[Fryingpan]

LB

Form Pro offers some thoughts and stats on this post

http://www.propun.com.au/forums/viewtopic.php?topic=1950&forum=7

[Luckyboy]

Fryingpan,

Many thanks for the prior post link. I was really interested in that % of winner in Top 5 Average Prizemoney.

Maybe that is my problem. Instead of looking for the Top 6 I should limit it to the Top 5. I'll run it over my data to see if it reduces the number of losing bets over the past few weeks.

Chrome Prince,

Many thanks for your insight. I thought the idea from 'formpro' reinforced your thoughts.

I have found a useful filter for identifying early career winners is to ensure that the career win strike rate is reflected or bettered in the horses past five starts.

This simple rule generally eliminates early achievers at 2 year old level that fail to go on.


LB

[xanadu]

G'day LuckyBoy,

I would like to offer my advice/opinion as I believe that prizemoney factor can be very misleading, as outlined by other learned contributors. It is quite common for owners of ordinary horses to accept for races in which their horse has no realistic chance. They hope to run in the first five to receive very attractive prizemoney(although their horse may be beaten by up to 20 lengths). This distorts the real class of their horse and a form student must take this into account when assessing the all factors.
To assist you in refining the chances in any race I suggest looking at the following factors:
1) only consider any horse finishing within two lengths of the winner last start(no matter where it finished);
2) look for a horse down in class and field strength from that run;
3) look for a replacement jockey, especially if a well-credentialled jockey replaces an apprentice.
I am confident that these factors(filters)will prove more beneficial than the prizemoney factor).

I hope this may be of some assistance.

Cheers.

[Luckyboy]

Xanadu,

Many thanks for your advice. I'm intrigued with your replacement jockey strategy.

Do you have any statistics to support this?

I have a two phased approach to bet selection.

Firstly, I have a statistical based system that narrows down the field size and then a qualitative based system accounting for other factors.

Here's my stat system:
1. Identify Top 6 Horses as per Friday paper prices;
2. Identify all horses ranked within four points of the top rated Zipform (Sportsman) horse;
3. Identify the Top Six Horses based on AVP.

In my analysis I have found these three filters have the highest correlation and statistically provide over 70% of winners.

I should add that I have program that pro-rates to a percentage, points 1 and 3.

Horses that meet the three above filters progress to a qualitative stage that scores each horse based on:

1. Last three start form;
2. Place %
3. Jockey
4. Barrier
5. Weight

Interestingly I do not look at distance. My analysis over 1500 races identifies distance as a 50/50 proposition. I tend to find a horse that is good at a 'distance' is easily recognised.

As I said in my original post, I was questioning the Top 6 AVP as a recent run of ordinary results was highlighted by a decrease in winners being from the Top 6 AVP.

My hypothesis is still inconclusive!


LB

[xanadu]

No, I don't have any stats as it is more a case of observation over many years. All the top stables do it and most of the top jockeys now have managers to procure rides for them, so they are on the lookout for prospective city winners.
You can see the strategy-an apprentice or unfashionable rider gives the horse a good hit-out then when the horse is primed a perceived "more competent/fashionable" jockey is engaged. This strategy has worked at the provincials for years and will continue to do so as most of the top hoops do not frequent the provincials regularly. If they do, especially for only one ride, it is time to take notice. One of the top hoops is only engaged when a horse is considered good enough to be brought to town. It is just another factor in your form study and I just offer it for your consideration.

Cheers.

[Luckyboy]

Xanadu,

A good logical thought process that sometimes is so obvious a lot of punters overlook it!

I appreciate your advice and I am always on the look out for assistance!


LB

[xanadu]

Luckyboy,

I was perusing your previous posts and one comment "rang alarm bells" for me. You stated that you rely on the newspaper prepost markets as one of your selection filters. In one word, "DON'T."
The reason is that to the best of my knowledge,these markets are prepared by a journalist, on the orders of his/her sub-editor, to provide a market in the vicinity of 135-145%, and let's face it, some of these "journalists," wouldn't know which end of a horse to feed!
Therefore, I suggest that you avail yourself of some expertise provided by some contributors to this forum.
Contributors like Form-Pro and TESTAROSSA have provided some very accurate ratings/pricelines and I am sure that if you based your ideas around these and some other contributors(sorry, I haven't got space to name them all), your results would improve immediately.
I hope this may have been of some assistance.

Cheers.

[xanadu]

Luckyboy,

In relation to my previous comment about change of "superior" jockey, I hope you noticed SR6 6.4.03 No 6 Star Of Florida ridden by Patrick Payne, replacing an accomplished jockey, R. M. Quinn-it ran second last start beaten 1.3lengths, and dropped in weight. It fitted almost all the criteria I outlined in recent posts and proved a profitable return for me-I hope you were on.(Although, after Payne's ride on Northerly yesterday, we would be very disappointed).

Cheers.

[Luckyboy]

Xanadu,

Appreciate your advice. My use of pre-post prices is not from a value perspective, but a rating perspective.

I have found this filter to be a very good field 'culling' tool. In my analysis of the past year and half, I have found over 70% of winners come from the Top 6 pre-post rated horses.

LB

PS: Thanks for the jockey tip. It came to good effect today!

[xanadu]

Luckyboy,

I might be able to assist you with improving your refining of chances in pre-post markets. You state that newspaper pre-polls secure approximately 70% of winners by refining your analysis to the first six horses in the market. As I have said previously, these markets are not a true reflection of the market.
To alleviate this I suggest you certainly refine your study to the first six in the market but make sure you use the opening market quoted on the radio or preferably the opening market shown on Skychannel.
These are infinitely more accurate as they take into account overnight market movers and support for first starters etc
I am certain if you utilise this strategy you can refine your success rate to about 85%(only taking notice of first six in market).

Cheers and let me know how you get on.

[xanadu]

Luckyboy,

In relation to that jockey replacement factor, it works a treat with Greg Ryan around the provincials. I have been researching this approach for a while and I believe there may be the makings of a system there, concentrating solely on G M Ryan's mounts. The only draw-back is that rank and file punters follow him religiously and sometimes value is hard to get but it provides an excellent strike-rate.

Cheers.

[Luckyboy]

Xanadu,

Thanks for the opening price tip. I will have a look at this over the next few months and give you an update.

I agree with you about Greg Ryan. The challenge is to find value with him. On the subject of jockeys (and I have never actually annalysed it) but Nash Rawiller is one who springs to mind particularly with horses trained by Mayfield-Smith.

You might have some information on this?


LB

[Chrome Prince]

Luckyboy,

After running a few tests regarding Average Prizemoney, the filters which have most impact are Win % and last start lengths beaten or won.

I found that you can turn a level stakes profit backing horses starting at greater than $2.00 but less than $3.00, ranked first in API, greater than 50% win rate or greater and last start winner.
This was enhanced eliminating all 3YO horses.

This also showed a profit place betting.

It reflects that using API as an indicator along with other filters is very powerful.

Interesting to note that barrier position, track or distance win and place percentage, even weight carried was largely irrelevant to profit.

[Luckyboy]

Chrome Prince,

Thank you for your analysis. Can you tell me the number of bets and time period of your study?


LB

[Chrome Prince]

Luckyboy,

Sorry for the delay in my reply.

This is the problem, my database was only started since 1/1/2003, so the number of qualifying bets is obviously minimal.

However, I do have confidence in the longterm success of it, because of the very high strike rate and the number of winners which make up the actual profit.

Here is a full workout of results....

RULES
1. Greater than or equal to 50% Win strike rate.
2. Ranked #1 Average Prizemoney.
3. Last Start Winner
4. Starting within 30 days of last start.
5. Tote price must be less than $3.00 just prior to the jump.
6. Meropolitan tracks only***

Bets - 50
Winners - 26
Placngs - 46***
Win Strike Rate - 52.00%
Place Strike Rate - 92.00%
Win Profit - $6.30
Place Profit - $9.00
Win Profit On Turnover 12.60%
Place Profit On Turnover 18.00%
Win Average Dividend $2.17
Place Average Dividend $1.28

If you eliminated all horses starting at less than $2.00 this is the result....

Bets - 31
Winners - 17
Placngs - 27***
Win Strike Rate - 54.84%
Place Strike Rate - 87.10%
Win Profit - $11.20
Place Profit - $5.90
Win Profit On Turnover 36.13%
Place Profit On Turnover 19.03%
Win Average Dividend $2.48
Place Average Dividend $1.37

Obviously, the sample is small, and needs a lot more data to draw conclusions, but it does indicate the power of strike rate combined with API.

What gives a lot of confidence is the very high place P.O.T. provided you stick to the price filters.

I have followed this for some time, and while sometimes you'll miss a borderline bet because of very late price fluctuations on the tote, this is more than compensated for by using IAS or Betfair to get much better dividends.

Hope this helps.

[Luckyboy]

Chrome Prince,

Very promising!

Given my recent poor run and you recent data is it possible for you to tell me the number of winners/placegetters if you changed the win % greater than 50% to a place % greater than 50%?

Sorry to appear pushy, but you got me curious!


LB

[Chrome Prince]

On 2003-04-08 20:03, Luckyboy wrote:
Chrome Prince,

Very promising!

Given my recent poor run and you recent data is it possible for you to tell me the number of winners/placegetters if you changed the win % greater than 50% to a place % greater than 50%?

Sorry to appear pushy, but you got me curious!


LB



No problem Luckyboy...I started the ball rolling, so I guess I have to follow up rather than leave anyone in the dark.

RULES
1. Greater than or equal to 50% Place strike rate.
2. Ranked #1 Average Prizemoney.
3. Last Start Winner
4. Starting within 30 days of last start.
5. Tote price must be less than $3.00 just prior to the jump.
6. Meropolitan tracks only***

Bets - 79
Winners - 34
Placngs - 69
Win Strike Rate - 43.04%
Place Strike Rate - 87.34%
Win Profit - <font color=red>-$6.20</font>
Place Profit - $5.50
Win Profit On Turnover <font color=red>-7.85%</font>
Place Profit On Turnover 6.96%
Win Average Dividend $2.14
Place Average Dividend $1.22

If you eliminated all horses starting at less than $2.00 this is the result....

Bets - 53
Winners - 21
Placings - 43
Win Strike Rate - 39.62%
Place Strike Rate - 81.13%
Win Profit - <font color=red>-$0.70</font>
Place Profit - $2.00
Win Profit On Turnover <font color=red>-1.32%</font>
Place Profit On Turnover 3.77%
Win Average Dividend $2.49
Place Average Dividend $1.28


Not a good result!

[Luckyboy]

Chrome Prince,

Thanks for your effort in putting together that piece of analysis.

Your findings have given me a bit more of an idea as to what may be wrong in my selection process - 'place percentage'.

I have found that you cannot look at place percentage in isolation of win percentage.

I think I may be on the way back!


LB

[Chrome Prince]

On 2003-04-10 09:53, Luckyboy wrote:
Chrome Prince,

Thanks for your effort in putting together that piece of analysis.

Your findings have given me a bit more of an idea as to what may be wrong in my selection process - 'place percentage'.

I have found that you cannot look at place percentage in isolation of win percentage.

I think I may be on the way back!


LB



Luckyboy,

A piece of advice.....

In THIS scenario backing very short priced horses (usually selective favourites) the win % filter created the profits but did not really increase the strike rate by much.

In other areas it may be the place % which is a better filter....2nd 3rd faves etc.

It is not necessarily the strike rate which creates the profit.

My advice is don't immediately think that win% is the be all and end all - it just created a profit in this example....but in others may not as you lose the likelihood of better average dividend.

[Luckyboy]

Chrome Prince,

Thanks for the friendly advice.

My own analysis has provided a pretty good direction concerning the merits of using win % as a filter if combined with current form (last five starts).

The correlation is between the career win strike rate and the win strike rate over the last five starts.

I am now combining in place % to refine it further making it more robust.


LB

[Chrome Prince]

Sounds good.

Actually I have been looking at similar theories

Applying Malcolm Knowles Power Of Ten staking system theory to that of Horse Form.

So it's not the staking plan I'm looking at, rather the theory behind it, to rate recent form.

[Bhagwan]

I ran this through the last 10 mnths.

RULES
Ranked No.1 Av.Prz. Money
Won Last Start
Win SR of 51-100%
Pre-Post Newspaper price $3.10-5.50
Must be TAB 1st or 2nd Fav 1 min before jump, regardless of it`s TAB price.

Metro races , any day of the week.
46.7% SR win
34% POT

Sat only Metro
50% SR
60% POT
Approx 8 bets a month, if betting Sat only.

[Chrome Prince]

On 2003-04-11 07:19, Bhagwan wrote:
I ran this through the last 10 mnths.

RULES
Ranked No.1 Av.Prz. Money
Won Last Start
Win SR of 51-100%
Pre-Post Newspaper price $3.10-5.50
Must be TAB 1st or 2nd Fav 1 min before jump, regardless of it`s TAB price.

Metro races , any day of the week.
46.7% SR win
34% POT

Sat only Metro
50% SR
60% POT
Approx 8 bets a month, if betting Sat only.




60% P.O.T. ????

Bhagwan, my figures are massively different to yours.

Based on your rules...

Sat Metro
selections 155
Winners 56
Strike Rate 36.13%
Profit <font colour=red>-$.30</font>
Profit On Turnover <font colour=red>-
0.19%</font>

Am I missing something?

:eek:

[Bhagwan]

I forgot to click my following screen, I`ll try again.

RULES
1)Metro Sat S,M,B,A,P
2)51%-100% Win SR
3)Pre-Post newspaper market $3.10-5.50
4)Won last start
5)Going- Fast, Good, Dead.
6)Last start 5-75 days
7)Barriers 1-11
:cool:Hcp Wt. 58kg.
9)Actual Wt. carried 58kg.
10)Field size 8-14
11)Has to be 1st. or 2nd TAB Favs. 1 min before jump.

50% SR
60% POT
Approx 4-5 bets a month

[Bhagwan]

That weight factor should read 58kg & less.

[xanadu]

Luckyboy,

In respect of the jockey change, I hope you noticed the strategic change of jockey in Melb Race8 No 3 Sound Action,
previously ridden by very competent jockey, W Hokai, but replaced with L. Currie on this occasion.
It duly won, paying $2.80w $1.60pl but I was on and rec'd a good dividend.
The astute punter does not mis these opportunities.

Cheers.

[xanadu]

Luckyboy,

I spoke to you recently about the strategic change of jockey steering us onto a potential top winning chance. I think I have identified one for you today:
SR3 No 1 Sea Jester
it has been well ridden by L Moloney in it's last two wins and he is now replaced by "the pumper" J. Cassidy- I would expect a bold showing.
Also, keep you eyes out for a runner at Echuca Race4 No 2 Ruby's Jester, scratched the other day to run at a minor meeting-we'll see how it goes.

[xanadu]

I suppose the jockey replacement rule could also apply to :
SR3 No 9 Laminated
where L Beasley has replaced R Spokes, we'll se how it goes.

Cheers.

[xanadu]

I hope you got on!
SR3 No1 Sea Jester 2nd $3.10pl
a very profitable race, even if Laminated was unplaced.

Cheers.

[xanadu]

Luckyboy,

I hope you got on:
Echuca race4 No 2 Ruby's Jester 1st $4.40 and $1.20pl

Cheers.

[Chrome Prince]

Xanadu,

Well done!

Certainly was a good thing and won with a fair bit in hand.

:wink:

[xanadu]

Thanks Chrome Prince,

They are out there(ie good priced winners) if you are prepared to be selective and wait for the right time.
We'll see what the rest of the week has to offer.

Cheers.

[Luckyboy]

Xanadu,

Sorry I have been AWOL for the past week. I did pick up Sea Jester and Laminated, but missed the Euchuca race.

The jockey change is a good little system. One thing I do like to check with it (via DFS) is if whether the jockey has ridden the horse before - this is an even better pointer.


LB

[xanadu]

Yes, I agree Luckyboy, it is a good aspect to take into account.
There are some interesting things happening today at Hawkesbury:
SR5 No1 Arbitrage
-scratched from Canterbury yesterday. Also, unfortunately, in the same race:
SR5 No 4 Euripedes
-Beadman's only ride for the day(remember my comments about leading jockeys going to the provincials for only one ride?
SR7 No 1 Mining Silver
-scratched from Monday to run here in an easier race.
These are pointers to help us determine what stable intentions are-we'll see how they go and hopefully return us a profit.
Unfortunately, the change of jockey theory let me down yesterday at Canterbury:
23.4.03 SR5 No 1 Youthful Edition
won it's previous two races in the country. It was brought to town, a capable apprentice replaced it's regular jockey and it had the advantage of a 3kg claim. Unfortunately, it was not up to the task but that's the way it goes. I went against one of my own rules here and backed a horse ridden by a 3kg apprentice but I thought it had enough pluses to outweigh the fact it was to be ridden by a 3kg claiming boy but it couldn't do it.

Cheers.

[xanadu]

Luckyboy,

I hope you got on!
SR5 No 1 Arbitrage unp
SR5 No 4 Euripedes WON $2.0w $1.40pl
SR7 No 1 Mining Silver WON $1.8w $1.10pl
not great prices but a good result anyway.

Cheers.

[Luckyboy]

Xanadu,

I couldn't get on today. I had to attend to some important family matters - take the wife out to lunch. Allows me to get a 'punting' grant for the next Saturday.

Now, a quick one to think about and noting your penchant for jockeys! If you go through the typical Jockey Premiership list you will note that a small number of jockeys make up the greater percentage of winners.

For Saturdays only, use these jockeys and identify their best ride each. You'll be very surprised by the strike rate!


LB

<font size=-1>[ This Message was edited by: Luckyboy on 2003-04-24 21:40 ]</font>

[xanadu]

Yes, you are right Luckyboy and it is something which I take notice of, especially if one of the top hoops is replacing a perceived "less fashionable " jockey. For instance,today, there is an example:
MR5 No1 Glefti
top jockey, S. King, replaces apprentice, Michelle Payne-expect a good run.
Other potential winners today are:
Gosford R3 No2 Don't Tell Me Lies* scratched from Sydney to run here.
Adelaide R2 No12 Entice*reasonable run last start and drops back in class, from the strong McEvoy stable and ridden by top jockey, S Price.
Good Luck.

Cheers.

[xanadu]

I hope you a good result from these:
Gosford Race3 Don't Tell Me Lies 2nd $1.60pl
Adelaide Race 2 Entice Won $2.30w $1.30pl
Melb Race 5 Glefti 3rd $1.60pl*
*what a race!......Glefti was in front in the last 150m but was swamped and held on for 3rd place. POT=36.7% - a pretty good result considering we were a little unlucky in a couple of the races.

Cheers.