Do I have a genuine edge or is this just statistical variance?

[psychoff]

Experimenting a single strategy in over 4000 events and saying that you are in profit means, you have an edge for sure. You may start to pray for it lasts long and smash as harder as you can before it vanishes. Welldone!

[Atho55]

This strategy as paper tested was last mentioned at the time of the Cheltenham festival and fits with the topic of this thread. When runners in the race are either 5,6,10,11,14 or 18 then Back Rank (BSP Fav) 4,5,7,8,10,11 or 16. BSP < 100. Stake is £25

Since 1st Feb when the data collection started it has been a roller coaster of ups and downs but I believe the ups can be predicted.

These are the results I can screen grab...

Strat A.jpg

Of late it seems to have been doing rather well. I think it does better when there is a prestige meeting amongst the cards either by horses that would normally be at the lower meetings being entered at the prestige meeting thus leaving opportunities or being specifically aimed to do well by whatever means at the main event.

This is not a get rich quick scheme by any means despite the bottom line figure being attractive, it had a run of just under -5k at one point hence the emphasis on the entry point to start.

I am looking to agree with the opening post that within the trading options available statistics do look to have a part to play

[LinusP]

Do you think a strategy randomly based on the BSP is going to be profitable long term?

[Anbell]

If you think have an edge and given your mathematical brain skip over here - https://stattrek.com/online-calculator/binomial.aspx

gazuty, you often point people to this link (thanks!) Can I dare ask for a mini-tutorial about how to use it?

For example, I've just plugged in some numbers from one of my bots. I use 100/(median odds) = 0.167 (approx 6.0) over 80 trials, and 22 winning bets (successes)

So, based on market price, I should win 17% of the time, and I've actually won 27.5% of the time. (this is pre-off betting, not trading)

The Binomial probability is 0.00526065831
Cumulative probability (P(X < x)) is 0.98987105591
Cumulative probability: P(X < x) is 0.99513171422

Could you please explain what these numbers mean? (I presume it doesnt mean that there's less than 0.5% chance that those results are edge)

Thanks so much

[gazuty]

If you think have an edge and given your mathematical brain skip over here - https://stattrek.com/online-calculator/binomial.aspx

gazuty, you often point people to this link (thanks!) Can I dare ask for a mini-tutorial about how to use it?

For example, I've just plugged in some numbers from one of my bots. I use 100/(median odds) = 0.167 (approx 6.0) over 80 trials, and 22 winning bets (successes)

So, based on market price, I should win 17% of the time, and I've actually won 27.5% of the time. (this is pre-off betting, not trading)

The Binomial probability is 0.00526065831
Cumulative probability (P(X < x)) is 0.98987105591
Cumulative probability: P(X < x) is 0.99513171422

Could you please explain what these numbers mean? (I presume it doesnt mean that there's less than 0.5% chance that those results are edge)

Thanks so much

So what the results are telling you is that your probability of getting exactly 22 winning results is approximately half a percent or 0.526%.

Your probability of getting 21 or fewer winning results was 98.987%.

Your probability of getting 22 or fewer winning results was 99.5%.

Your probability of getting 23 or more winning results was 0.499%

Your probability of getting 22 or more winning results was 1%.

It looks likely you have an edge. The less likely your result or your result plus more wins the more likely you have an edge. You need to have sufficient trials and a low percentage of just random luck.

[Anbell]

So what the results are telling you is that your probability of getting exactly 22 winning results is approximately half a percent or 0.526%.

Your probability of getting 21 or fewer winning results was 98.987%.

Your probability of getting 22 or fewer winning results was 99.5%.

Your probability of getting 23 or more winning results was 0.499%

Your probability of getting 22 or more winning results was 1%.

It looks likely you have an edge. The less likely your result or your result plus more wins the more likely you have an edge. You need to have sufficient trials and a low percentage of just random luck.

Thanks. That's really interesting.

With such a small number of trials I'd have thought that the probability was closer to 70%, than 99%.

Thanks for your help.