Cut Your Losses?? (Mean Reversion)

Learn sports betting strategies and discuss key factors to consider when placing a bet.
Post Reply
PeterLe
Posts: 3715
Joined: Wed Apr 15, 2009 3:19 pm

Hi
I dont know if anyone would find this of any interest..but I'll share it anyway in the event that it may be..

Today I was running some tests (on Auto) on the market on three totally separate accounts (Identical markets; Clocks totally in sync etc...). This would then serve to feed into a another system I'm working on.
I was trying to determine the effect stop losses (Cutting your losses) would have on the days P&L...(and mean reversion). All races were pre play (6 Min prior to the off, and greened/red up pre inplay, low stakes)
The system would wait for the offset to be matched before placing another bet (just to stop it running away with itself with me not being there)

Account One
Looking for one tick scalps
No Stop Loss
Matched 278 Bets
Total amount staked £4,860.06
Day P&L: -£1.43

Account Two
Looking for One tick scalps
Stop Loss at 4 Ticks (placed at 5)
Matched 343 Bets
Total amount staked £5,779.31
Day P&L: -£15.59

These two accounts above were the ones I wanted to compare and I expected account two to be better to be honest. So I ran a third account to compare against account two:-

Account Three
Looking for two tick scalps
Stop Loss at 4 ticks (placed at 5)
Matched 292 bets
Total amount staked £5,113.63
Day P&L: -£19.24

So in Summary; it seems that cutting your losses doesn't always pay dividends?

Looking at account one; It's amazing how efficient the markets are.
Anyway the strategy used in account one will be the one I'll use going forward, yes there will be days where there will be losses with this, but over the course of a month it should even itself out...

Anyway, make of it as you will, hope it helps in some way?
Thanks
Regards
Peter

(I also seem to remember reading in trading in the zone (and in Peter's blog) how some people often sell just as the price is peaking and buy just as the price reaches a low. It seems that Mean reversion is strong in horses racing markets - overall?

It would also be great if BA could calculate the average price traded (VWAP ?). I cant do this as I usually tend to switch to the market just pre inplay. It would be great if it was possible to drag this data from the BA Server??
User avatar
LeTiss
Posts: 5386
Joined: Fri May 08, 2009 6:04 pm

I never use the stop loss facility anymore Peter

In principle it sounds perfect, in reality using stop losses is the easiest way to achieve a disastrous result.

There will quite often be momentary gaps in the market. Stop losses just close out for a loss that could easily have been avoided trading manually
User avatar
gutuami
Posts: 1858
Joined: Wed Apr 15, 2009 4:06 pm

I use stop loss only when I go away from my computer, usually for a coffee or tea.
User avatar
kinglouie
Posts: 81
Joined: Mon Jul 12, 2010 1:45 pm

Have to agree about auto Stop Losses, great idea in principle, bad idea in practice just because of how the liquidity in a market can change.

You can be in a good position, stop loss in place, market being stable or even moving in the right direction for you, then bang..............a red figure and your out for a loss.

Even with a trailing stop loss this still happens and is very frustrating!
rpadrela
Posts: 72
Joined: Sun Apr 25, 2010 9:19 pm

I've been trying to convince myself to use stop losses and cut my losses but I've been a bit skeptic precisely because of liquidity problems.

I totally agree with what you guys are saying.
User avatar
fink nottle
Posts: 20
Joined: Thu Nov 19, 2009 10:42 am

Apologies if I have read this wrong but in example 2 why would you have such a long stop if you were only scalping for a tick profit. Surely you then require a high strike rate to counter any losses?
PeterLe
Posts: 3715
Joined: Wed Apr 15, 2009 3:19 pm

fink nottle wrote:Apologies if I have read this wrong but in example 2 why would you have such a long stop if you were only scalping for a tick profit. Surely you then require a high strike rate to counter any losses?
Just merely comparing different strategies...quite often a price can come in a few ticks before going out again
booboo
Posts: 29
Joined: Wed Apr 28, 2010 1:55 pm

Hey there Peter.

Re "Just merely comparing different strategies...quite often a price can come in a few ticks before going out again"

Agree. But always use stop losses. More for "emotional" reasons. I never enter a trade without having an idea of where my loss "should" come in. Then I move it closer.
Your project is to look at "not getting stopped out" and that is subjective to the trade you enter.
Thisthen also brings in, where do you enter the trade...and so on.
If each trade is planned before hand, then an idea of the "risk" is taken into account. And so the stop.
If the "fundementals" of "why" you entered a trade still hold true, then re enter, however, if you have entered the trade...and have not taken into account the "risk" of that trade, then a "standard" 5 ticks out doesnt apply.
Dont forget, that on the opporsite end of the trade, is a "trader" or "punter"....and they dont operate to "standard" !
And....I am sure you know this...and do this anyway...a "perfect" trade, only has a small entry retrace anyway, if done right ! The "stop" is your safety belt ! And thats all it is !
If you get stopped out, you havent consider the "trade" correctly. re evaluate where it went wrong....or you have just saved yourself in a head on crash !

David.
booboo
Posts: 29
Joined: Wed Apr 28, 2010 1:55 pm

Oh...and as a P.S.

The guy that said "he only uses stops when he goes to make a cup of tea"....is incredibly frightening !

I guy with a "bag of sand" on the board, who then goes off to make a "cuppa" ...should seriously re consider what he is doing...or his need for a cup "o" tea !

David.
lewismbet
Posts: 55
Joined: Thu Jul 23, 2009 11:20 am

Really good info, I have a scalping bot which is running in profit and have often wondered how much I am losing by effectively 'crossing the spread' on my stoplosses particularly when edges can be very small and spikes/volatility can hit stops that you would not have closed manually. I may have to take the plunge into a stoploss free environment and accept the large swings which will inevitably appear however if it more +EV on a weekly or monthly scale count me in.

My only worry is that it may be possible to be seriously burned on one serious drifter/steamer without reversion. Will report back in the future if I find results.
Zenyatta
Posts: 1143
Joined: Thu Mar 11, 2010 4:17 pm

It's fascinating to watch the price chart of a selection on the few occasions when I'd known in advance exactly what the starting SP should be (due to detailed info of a horse's form).

The market does move to the correct SP by the off-time, but doesn't get there in a linear fashion. For starters, the period from 10 minutes-5 minutes before the off has a high degree of randomness, the market splutters around and the price can sometimes move 5 ticks in the wrong direction. It's only in the final 5 minutes that the price really starts to rapidly 'home in' on the correct SP, and even then there can be temporary spikes or fluctuations of 2 or 3 ticks in the wrong direction.
Iron
Posts: 6793
Joined: Fri Dec 11, 2009 10:51 pm

Hi Zenyatta

With steamers and drifters, do you find that, due to market hype, the price ends up much lower or higher respectively than its true price?

Jeff
Zenyatta wrote:It's fascinating to watch the price chart of a selection on the few occasions when I'd known in advance exactly what the starting SP should be (due to detailed info of a horse's form).
PeterLe
Posts: 3715
Joined: Wed Apr 15, 2009 3:19 pm

Deleted..
Last edited by PeterLe on Mon Aug 06, 2012 12:41 pm, edited 1 time in total.
lesblakeman
Posts: 15
Joined: Wed Jun 22, 2011 4:02 pm

After a couple of near career ending losses in May I decided to go the "stop loss" route , lasted about an hour , the non linear movement of the markets just makes it unworkable on horse racing .
Iron
Posts: 6793
Joined: Fri Dec 11, 2009 10:51 pm

I've found this thread very interesting.

Has anyone else experimented with random trading, where you enter at random with a stop of x ticks and an offset of y ticks?

Jeff
Post Reply

Return to “Betfair trading strategies”