Malcolm Knowles - Consistency book

[Barny]

Is the data in this book directed at only those horses with 20 or more starts ?

It's a good good read .... but I do have a couple of issues with it, which I'll raise a little later.

[Pauls123]

that's those little pocket size books..?,.....I've got a few of them laying around somewhere, read them years ago,

Paul

[Barny]

Yup, there the ones. Full of typos, grammar mistakes etc, etc .... but a good read BTW.

[gunny72]

I have studied these books in detail, particularly the consistency one as I found this to be the only one that is relevant. Also, I had been using place percent reasonably successfully for several years before this book came out.

I simply backed the two longest priced horses from the top three place percent rankings and regularly obtained brilliant results. I have found his percentages on place percent rankings have stood up for about 20 years and still do but, and not surprisingly, his stated and my profit on turnover results have deteriorated significantly over recent years. Whereas $50 to $80 divvies once were common, now this range seems to be $10 to $30, and this is not enough for my approach.

I believe that he meant there had to be 20 starts and I took this into consideration initially and missed out on many good winners so I forget about the minimum 20 starts and it made no difference.

The suggested filtering ideas in the book did not work for me. I found the best filter to be price.

The thrust of his consistency book though is to provide an improved method of adjusting DS ratings for consistency and is probably this is a good use.

Gunny72

[Barny]

The Win SR rank and the Place SR rank are even when the win and 2nd are added. The whole book ispremised on the outstanding Place SR ranking, and fair enough given the stats put forward, but the alarm bells ring when the 2nd place is taken into account.

No mention of how many races has been analysed.

And the analysis seesm only to take into account 20 starts or more.

A good read though, but further proof about how potentially MISLEADING STATS can be.

[Barny]

The DUAL winners and the LS winner confirms much commentary on here, and the stupidity of the public is also mentioned.

A good read ..... take nothing for granted unless you are quite happy to donate your pay to the TAB.

[gunny72]

Knowles says that place percent is the single best factor he has found and I suggest that it is foolhardy to ignore it.

[darkydog2002]

I found too that Malcolms Weight stats were most excellent - The main thrust being that those weighted 55 KG + accounted for 36.5 % of races yet won 50.6 %..
Info like this is too valuable to ignore.

Cheers
darky

[Bhagwan]

The place ranking stat is excellent.

There is only 2 other ranking methods that beat it slightly.

Thats is....
Average Prze money (even on Maiden events , surprisingly)
&
Pre-Post market ranking.

We are only talking about 1-2% difference.

[Barny]

Bhagwan this is a copy post on here ..... I don't have the time to search to verify that it's legit for you so take it as it is please.

1) Horse must have between 80% and 99% place strike rate

2) Must be the only selection in the race

3) Must have had at least 5 starts


5063 selections

Win return $4,588.20
Place return $4,474.28

So it's a 9.4% loss on turnover approximately.

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" /><o:p></o:p>

That's as per your rules taking out multiple runners in a race.

<o:p></o:p>

I quickly double checked it back to 2001 and out of 4424 races there were only 967 winners. A miserable 21.84%.

Although it's not clear, Malcolm Knowles stats seem to come from those that have had 20 starts or more. Malcolm Knowles suggests an 8% POT whereas these stats show a 9.4% LOT. That's a massive difference ..... some would say it's a 1.4% difference, some would say its.17.5% difference using 8% as the anchor and some would say it's a drop of 14.8% from the 9.4%. Malcolm Knowles stats on the Place% show no advantage over the Win% when the results of 1st AND 2nd are taken into account. Taking only 1st, then the stats are significantly in favour of the Place% over the Win% as is suggested in the book.

98% of all putts, if not hit hard enough (no matter how accurate they are !) won't drop into the cup ! I've also heard in exasperation from a golfer "Damn !!!! I didn't hit it !!!" ..... How the heck did it get there then ???

Happy to be of assistance ..........

[gunny72]

Knowles' results use place % ranking and not absolute place percent. The top ranked horse could have just 30% places and still be his no. 1 selection for a race.

I agree that the top ranked horse nowadays shows about a 10% LOT but the top three ranked place % horses still produce close to 48% of Saturday winners and my records indicate this has been the case since 1992.

I think more punters are using these stats these days as once I could make a modest profit using the two longest priced starters (3 mins to go on the tab) of the top 3 ranked place% horses but not lately.

[Barny]

gunny72, yup top ranked place % ..... wrong tram again for me !

Thanks for that !

Malcolm Knowles book put too much emphasis on the top ranked place % compared with the top ranked win %. Take the top TWO in both the win % and the place % and it's line ball. The figures seemed to cute for mine and it's interesting that things have changed as you said.