Laying - Level Stakes vs Liability

[The Ocho]

Hi all,

Just got a hypothetical question for you guys.

I'm just looking at something which involves laying certain horses in the price bracket of $2 to $3 on Betfair.

What would be the best way to do this - laying at level stakes or laying to liability?

I cannot give you the stats or percentage of winners/losers involved because I don't know myself (maybe someone could help with this). I'm just wanting to know what may be the best way to go.

Taking commission out of the equation for now and just using simple numbers we get the following:

Laying minimum $5 level stakes
Price $2.00 - Lay win $5 - Lay loss $5
Price $2.50 - Lay win $5 - Lay loss $7.50
Price $3.00 - Lay win $5 - Lay loss $10

Laying say to $10 Liability (so minimum $5 lay bet requirement is met)
Price $2.00 - Lay win $10 - Lay loss $10
Price $2.50 - Lay win $7.50 - Lay loss $10
Price $3.00 - Lay win $5 - Lay loss $10

Is there really a difference? Level stakes means you lose less on the lower priced horses which are expected to win more often HOWEVER liability means you win more when a lower priced horse wins.

Is one better than the other?

[SpeedyBen]

Ocho
Ideally with laying you want to win the same on your longer priced selns as you do on the shorter ones because they are going to lose ( win for you ) more often, statistically speaking. So, in my actual experience, I would always have won more laying a stake not to a liability ( I have done lots of xls studies to prove this ). However, since I like laying longshots I use the liability method because I don't want to wake up in the morning having laid a 330/1 winner. In practical terms the worst loser I have laid in the UK in the last 6 months was about 33/1 so I wouldn't have woken to a nightmare.
With the low price range you mention I reckon you will be slightly better off laying a stake but in that narrow range there should not be much of a difference.

[Chrome Prince]

It depends on the edge.
Is your edge fluent across all odds ranges or is the edge towards the shorter or longer end?
Laying to fixed stake can be better, but laying to liability can also be better as it magnifies your edge if it's towards the shorter end.
The simple answer is, it is not the same.
It will be different one way or another.

[The Ocho]

Thanks guys. I've just broken the odds range down to 10c increments with the following results but I had been using a $30 liability for testing.

2.00-2.10.....-13.78
2.12-2.20.....193.26
2.22-2.30.....163.16
2.32-2.40.....-3.72
2.42-2.50.....144.48
2.52-2.60.....204.56
2.62-2.70.....120.96
2.72-2.80.....56.34
2.82-2.90.....76.75
2.92-3.00.....-56.59

I've now changed the last increment up to just 2.92 (back fitting or what) giving me.....46.86 (instead of -56.59).

That's +988 for laying with $30 liability (no commission deducted though).

It is only 134 bets though however so it's probably running really hot (or cold as it's laying) right now.

There have been 34 losing lay bets out of those 134 races to date (75% SR).

Do the above figures give anyone any idea if level stakes or liability would be better?

[Chrome Prince]

Yes 134 bets isn't enough, but at this stage you're better off laying to a fixed stake.

[The Ocho]

Thanks very much Chrome Prince.

[The Ocho]

I just checked my figures (pre-commission let me stress) for those 134 bets

If laying to $5 level stakes +247
If laying to $10 liability +329

Today there were 11 races with 7 winning lays
$5 level +2.51 (only +$0.23 after commission)
$10 liability +9.76 (+$6.52 after commission)

I still don't believe I am comparing apples with apples so I suppose it's just a matter of which side of $2.50 the winners/losers fall for which way would be best.

[Chrome Prince]

Ocho,
I must say I'm very surprised by those results given the previous figures you posted. Perhaps what has a further bearing is the number of bets less than $2.50 and greater than $2.50.

[The Ocho]

Ocho,
I must say I'm very surprised by those results given the previous figures you posted. Perhaps what has a further bearing is the number of bets less than $2.50 and greater than $2.50.
I was a bit surprised too with those figures for yesterday. Obviously the number of losing bets matters a lot. For instance with the $5 level stakes if all 11 races won you would have +$55 (less commission). Lose a race and you lose the $5 while also paying out between $5 and $10 (depending on odds). So probably not too surprised actually.

[The Ocho]

Yesterdays results were a bit better with the system picking 13 races with 10 winning races and 3 losing races (77% SR - all picks between $2 & $2.92). The losses were at odds of 2.88, 2.66 & 2.56. The winning lays ranged from odds of 2.02 to 2.86.

Saturday 09/02/2013
$5 level +21.26 (after commission)
$10 liability +36.72 (after commission)

As mentioned in my trading thread, I may put $300 aside to give this one a go. If I do I think I will use the liability betting.

[The Ocho]

I have been trialing this with real $5 lay bets since only last Wednesday 6 Feb.

So far there have been 39 bets for 30 winning lays and 9 losing lays giving a return of $67.09 after commission of 6.5% (76.9% SR). TheSchmile's Gretchen was a lay pick today. ;) So far with my testing there have been 170 bets with 43 losing lays (74.7% SR).

I will place $200 aside for this and use the $5 level stakes method but I will use a percentage of bank rather than pure level stakes. This will be 2.5% of the $200 bank so starting at $5 lay bet. The bet amount will then increase if the bank increases and stay at that higher level should there be any loses (ratchet staking). This started today with 3 successful lay bets.

[The Ocho]

By the way, does anyone know of the strike rates per price in the $2 to $3 price range laying on betfair?

Just round figures:
I assume a $2 horse would win 50% of the time.
Would a $3 horse win 25% of the time?
What about a $2.50 horse? Would they split the difference and win 37.5% of the time (round figures)?

I'm just wondering.