Hi All,

I've been backtesting a lay method using my ratings as the mechanism helping me decide when to lay the favourite due to the race being open in the view of my ratings.

I'm just interested in what are acceptable stats for a lay method? I've only back tested for a week so far as I started last night but here's what i've got so far:

MAX LIABILITY & NO STOP LOSS (74 bets)
77.03% SR
10.70% POT

NO MAX LIABILITY & NO STOP LOSS (74 bets)
77.03% SR
8.95% POT

MAX LIABILITY & STOP LOSS (62 bets)
82.26% SR
17.10% POT

NO MAX LIABILITY & STOP LOSS (62 bets)
82.26% SR
16.18% POT

Obviously there isn't enough data to say whether its a winner yet or not but i'm just curious what makes a lay system acceptable? Definitely seems like managing the risk with a max liability and stop loss helps the returns. There has only been 1 occurance of 2 losses in a row so far and this would've been stopped due to the stop loss. So it may even be more profitable with some sort of recovery built in as well maybe?

Anyhow just looking for some comments on lay systems basically, if the backtesting keeps looking decent i might start live betting it with a bot with small amounts of money.

Cheers

Hi, no there's a lot more testing needed there. 74 is a very small sample. By comparison I'd probably lay twice as many horses as that a day

Obviously an acceptable lay system is one that gives you a profit and there's many way's that can be achieved. More targeted systems as you are alluding to do tend to produce a higher profit on turnover than high volume systems that make profit from the churn. I'm not averse to the more targeted approach but it can become frustrating when the Fates turn against you.



Maybe but too early to know. I'm not a stop loss person as I think it tends to even itself out over time. Stop loss can equal lost opportunity.



No no never do that! Silly for backing, suicide for laying.

__________________
Pixie

"It's worth remembering that profit isn't profit until it's spent off the racecourse." -- Crash

PS: How are you calculating your POT?

__________________
Pixie

"It's worth remembering that profit isn't profit until it's spent off the racecourse." -- Crash

Hi Angry Pixie,

The way i'm doing it is only laying 1 horse per race in selected races so some days there can be up to 10 or so selections, other days 1, 2 or 3. The recovery part was just an idea but one I don't really like, as you can read anywhere martingale type systems only work if you have a never ending bank account (according to statistics).

For the Max Liability POT i've calculated it as:

Profit / Total Liability to date

So for instance 5 lays:

Loss @ $3.50 odds - $10 Liability - $10 loss
Win @ $2.50 odds - $10 Liability - $6.67 Profit
Win @ $4.90 odds - $10 Liability - $2.56 Profit
Win @ $3.10 odds - $10 Liability - $4.76 Profit
Win @ $5.30 odds - $10 Liaiblity - $2.33 Profit

This would be a $6.32 Profit over $50 Liaibility, hence 12.64% POT, is this the correct way of doing it?

In terms of no max liability i've simply calculated based on $10 stakes so for the above example.

Loss @ $3.50 odds - $25 Liability - $25.00 loss
Win @ $2.50 odds - $15 Liability - $10.00 Profit
Win @ $4.90 odds - $39 Liability - $10.00 Profit
Win @ $3.10 odds - $21 Liability - $10.00 Profit
Win @ $5.30 odds - $43 Liaiblity - $10.00 Profit

This would be a $15.00 Profit over $143 Liability, hence 10.49% POT, is this correct?

I've done a little more testing, best result is still the Max Liability & Stop Loss option:

88 bets:
82.95% SR
16.87% POT

Some very sound advice by Pixie there. Stop / Loss isn't my cup of tea either.

I always calculate profit as a percentage of your total 'hold'. In your 2nd example, the $10 flat stake lays, your total hold is $50 obviously. Just remember to account for the 6.5% comission, so your profit was $9.35 x 4 less $25, $12.40 profit on $50 hold, thus 24.8%.

My preferred method of staking is simply 2 lay bets on each selection; one to Liability at SP and one to flat stake at available lay price at a pre-determined time, say -5 sec mark.

"My preferred method of staking is simply 2 lay bets on each selection; one to Liability at SP and one to flat stake at available lay price at a pre-determined time, say -5 sec mark."

Does that average out the prices much Raven?
I am looking for a way to 'flatten out the lumps' so to speak, as the ups and downs are a bit of a ride at times with this laying caper.

__________________
Jose'.

Nothing wrong with that. A much bigger shock to the system when the inevitable 2 or 3 losing lays in a row happens though.

As Raven pointed out, profit on hold is how many layers calculate their profits.

__________________
Pixie

"It's worth remembering that profit isn't profit until it's spent off the racecourse." -- Crash

Using the profit on hold method doesn't this over-estimate the POT as it's not considering what you've actually put at risk to receive that profit, its only considering the profit side of it?

I've continued doing some backtesting and almost done back to the start of May now. Once i've finished that if it still looks profitable, i'll decide which bot I want to start using and trial the method walking forward.