Quickest and easiest method I have found for testing if a system is luck or actually looks to have a edge is using this calculator here:

http://statpages.org/confint.html

This was shown to me by a previous member JFC.

Enter your stats into, and see what range is possible from your results so far.
Very handy.

Moeee, Bhagwan & Thorns, Thanks for your input it is really appreciated.

I will follow up and post the results. I do like the idea of 4-5 smaller groups and also the calculator should prove invaluable. Once again, Thanks for your input, and from what I have seen, it is these contributions that makes this forum such a success.

Hopefully at some stage I will be able to feed something of value back into the forum as well.

Fred

I thought this was interesting so I am sharing it with you ...
Variation: .50 cents to indicate a drift or firm.

Firming Drifting

ALL Horses:	1533	1717
ALL Wins:	87	102
ALL Profit:	-$ 867.30	$ 138.35
Strike Rate:	5.68%	5.94%
ROI	-11.32%	1.61%

Highest run of outs:
Firming: 84
Drifting: 71

I know its only after 10 days, but does indicate the smart money may not be that smart? .... Interesting - but I am not sure its a viable system !

Did not come out that well .. but the first column are the firming results and the second column the drifting columns ... hope you get the drift !

Below is the result based on 711 races using Betfair SP at 10 seconds before close and the final SP. It shows the No 1 ranked horse at the 10 second mark mainly retains that position at close (85%) and the other 14% comes from horses that were ranked either 2 or 3 at the 10 second mark. And true to form those 1st rank in both places win 32% of the time! ... Fred

10 Sec	Close	Number	%age	Won	%age
1	1	607	85%	196	32%
2	1	84	12%	14	17%
3	1	12	2%	2	17%
Other	1	8	1%	1	13%
	Total:	711		213	30%

By the end of the weekend (or shortly thereafter), I will have collected the results of around 900 races (Betafair SP).

As the results are not what I expected, I plan to stop gathering this data after the 900 is reached, however if anyone can give me a reason to continue collecting than I shall do so as it imposes virtually no overhead.

Also if anyone would like a copy of the data (in an excel spreadsheet) I am happy to provide.

If however you request and receive the data, I would hope that should you find anything worthwhile you would post your findings back on the forum.

... Fred

Thanks for sharing those findings.

I notice that there seems to be a 2 pt edge over average stats being reflected there for the 1st & 3rd fav.
Interesting stuff.


Av stat is
30% 1st Fav
20% 2nd Fav
15% 3rd fav.

__________________
Cheers.

This might be why the results your were expecting are not showing.
Comparing the last traded price would be much better, because you are currently comparing the predicted SP with the actual SP.
This in itself is flawed, because the predicted SP is based on a different formula to the actual SP, therefore any difference will be the calculation rather than the movement.

I would also set the seconds to more like 30 or 60 to get more of an indication.

__________________
RaceCensus - powerful system testing software.
Now with over 413,000 Metropolitan, Provincial and Country races!
http://www.propun.com.au/horse_raci...ng_systems.html
*RaceCensus now updated to 31/01/2025
Video overview of RaceCensus here:
http://www.youtube.com/watch?v=W821YP_b0Pg

Thanks Chrome. I just went through Betfairs SP calculations and Chrome is correct. This does answer a lot of questions.

What my table did prove is that Betfair's CALCULATION to predict the final SP is very good - and hence why there was such little movement!

I will do some adjustments and base my snapshots on the last traded price. I will probably do 3 snapshots - at 120, 30 and 10 for a few different comparisons. This may now give me the results I was expecting!

I will keep you informed.

... Fred

Hey mate I would love a copy of this table (email to max stuff 09 @ g mail . com) as I too have been tracking market prices prior to jump. Primarily 10mins, 5mins, 1mins, 0mins and then closed.

Have not quite worked out how to use this yet. I was using it to work out a strategy of backing the 3rd fav at the 5min to go mark based on some threads by Bhagwan but it is just not consistent enough for me.