Today's Horse Racing

[eightbo]

Was a nice surprise to finish the day and find I'd been positively affected by the Taunton incident where the jockey failed to weigh in, after having several "near-misses" of big profits throughout the day.

[eightbo]

Spent this afternoon working on a scalping bot, only got the lay side completed, its just placed 14 successful lay scalps on the same selection back to back. Here comes the next market, 28 unsuccessful lay scalps in row

nice one, 1T offset I presume? ...and are you using a stop?

I built one today as well. Seems like both sides are working correctly by end of the day although was taking steady losses with a 2T stop.
Added another layer of confirmation and prep'd a few variations of it to run side by side over next few days and see if any of them can eek out a profit.

[Crazyskier]

Spent this afternoon working on a scalping bot, only got the lay side completed, its just placed 14 successful lay scalps on the same selection back to back. Here comes the next market, 28 unsuccessful lay scalps in row

nice one, 1T offset I presume? ...and are you using a stop?

I built one today as well. Seems like both sides are working correctly by end of the day although was taking steady losses with a 2T stop.
Added another layer of confirmation and prep'd a few variations of it to run side by side over next few days and see if any of them can eek out a profit.

I'm never surprised how often any kind of stop loss results in overall losses (though it may minimise them). My most successful strategies rely on successful trades outweighing the inevitable many-tick losses, as time and time again whenever I implement a stop loss, it turns modest winning strategies into losers.

Interested to hear anyone else's experience with this...

CS

[eightbo]

when I look at some of the names of theses runners it does make me wonder sometimes, just got home from work, caught this & will pretty much do for the day


Easy As That.png

Have a good chuckle at the names that crop up. Here's a few that got me over the last few days:
  Wots The Wifi Code
  Tour De Pub
  Taipan (name of my local chinese takeout back home)
  Excelman
  Overpriced Mixer
  Dontdooddson

[jamesg46]

Spent this afternoon working on a scalping bot, only got the lay side completed, its just placed 14 successful lay scalps on the same selection back to back. Here comes the next market, 28 unsuccessful lay scalps in row

nice one, 1T offset I presume? ...and are you using a stop?

I built one today as well. Seems like both sides are working correctly by end of the day although was taking steady losses with a 2T stop.
Added another layer of confirmation and prep'd a few variations of it to run side by side over next few days and see if any of them can eek out a profit.

Yeah one tick, I didn't run it on the horses today, I shortened the time and ran it on the greyhounds. I was hoping to export the history lists that show the % of volume traded at prices against the selection price so that I could determine entering where there has been plenty of matched volume and is potential support. I ended up not figuring it out, even though I've exported tonnes of history lists in previous files.

I'll send it over, maybe you could take a look and see what you think. Its a wip atm but I dont mind getting some input if you're willing.

[jamesg46]

when I look at some of the names of theses runners it does make me wonder sometimes, just got home from work, caught this & will pretty much do for the day


Easy As That.png

Have a good chuckle at the names that crop up. Here's a few that got me over the last few days:
  Wots The Wifi Code
  Tour De Pub
  Taipan (name of my local chinese takeout back home)
  Excelman
  Overpriced Mixer
  Dontdooddson

It got off to a flyer on the gh too only to fizzle out this afternoon. It was at 75% sr on 50 markets (not sure about individual selections) but with the gh being so volatile my avg losses started to creep well above my avg gains. I tried to use book % conditions to tighten things up but it made things worse... haha, so now I dont know how it would have done because I've sat tinkering with it all day.

[eightbo]

Interested to hear anyone else's experience with this...

CS

I don't claim to be an expert on the matter but here's my thoughts

Your stop is a bet just like your entry and you have the potential to place it into the market with edge e.g. if you can accurately determine more times than not the price will go further away from the price where you'd be stopped out. Useful when your entry is based on a particular key level holding but not so much when your strategy is only focused on current price action.

Stops are hitting the offered price so you're giving away a bit of value every time vs. true price if you assume the midpoint is accurate on average.
I can't find the pic but I have an image somewhere comparing a strategy 1. hedging before start of race @ offered price; 2. hedging @ reverse offered price ; 3. no hedging at all and equity curve for 3 was great, 2 was slightly profitable and 1 was heavily negative consistently. And that's only ONE hedge firing each market. If you have a shitty stop and a high number of entries you could be exaggerating that effect massively (the bot I made today is probs doing this)

Today I've added the stop at first as a sort of handicap during testing and see if I can get breakeven or better then go from there with a few dif. variations. If I don't feel confident that my stop point has edge I'll just omit it from the start, always have the option to add a stop in if you observe something useful either in your data or just from watching it visually.

Mostly we're using stops to protect downside but if we size our stakes way down so your bank can take the natural variance/drawdown of the random outcomes without any risk of ruin then you can remove the stop and enjoy the higher edge. Think this is the best way to go then gradually up stakes over time as the strategy proves itself.

edit: found the pic showing offered prices (assumes offered exits are matched so not a true reflection of reality):

◘.png

[eightbo]

...
I didn't run it on the horses today, I shortened the time and ran it on the greyhounds. I was hoping to export the history lists that show the % of volume traded at prices against the selection price so that I could determine entering where there has been plenty of matched volume and is potential support. I ended up not figuring it out, even though I've exported tonnes of history lists in previous files.

I'll send it over, maybe you could take a look and see what you think. Its a wip atm but I dont mind getting some input if you're willing.

...

Got a little backlog of automation stuff I want to tick off the to-do list first however happy to talk about your bot in PMs.

I used to trade GH manually every day and I have to say that the levels are very weak there. Usually you want to be involved in price-creation as it happens. If a level holds it's usually because there's a good stack of existing cash offered there beforehand as price comes into the level rather than someone stepping in to defend a price. Not always the case of course and you have a higher chance of a level being defended in the Open Races but that was typically what I noticed.

[Crazyskier]

Interested to hear anyone else's experience with this...

CS

... if we size our stakes way down so your bank can take the natural variance/drawdown of the random outcomes without any risk of ruin then you can remove the stop and enjoy the higher edge. Think this is the best way to go then gradually up stakes over time as the strategy proves itself.

◘.png

This is precisely what I've found, time and time again, and how I've refined my strategies over several years now.

What I have learned is that horse races such as Maidens, Novices and INHFs have HUGE variance from a few hours out to BSP; for example horses priced around 10 often hit 5 before the off, and horses that start the day around 6s can go sub-3s! This 'turbulence' is something that the grp races and most handicaps seem to have far less of, and therefore they must be approached in very different ways to exploit the variances in odds movement.

In short, the typical odds movement of certain races mean that they are far better or worse suited to the use (or not) of stop loss; and the majority of my findings and endless A/B testing, overwhelmingly support the smaller stakes, high frequency trades, relying on the gains to outweigh the losses over a certain number of hours pre-race, rather than trying to somehow limit losses mid-trade.

Hope this makes sense. I'd love to hear from more 'players' on their experiences of stop loss usage and it's merits or lack of.

CS

[eightbo]

Cheers for sharing CS

[firlandsfarm]

Spent this afternoon working on a scalping bot, only got the lay side completed, its just placed 14 successful lay scalps on the same selection back to back. Here comes the next market, 28 unsuccessful lay scalps in row

nice one, 1T offset I presume? ...and are you using a stop?

I built one today as well. Seems like both sides are working correctly by end of the day although was taking steady losses with a 2T stop.
Added another layer of confirmation and prep'd a few variations of it to run side by side over next few days and see if any of them can eek out a profit.

I'm never surprised how often any kind of stop loss results in overall losses (though it may minimise them). My most successful strategies rely on successful trades outweighing the inevitable many-tick losses, as time and time again whenever I implement a stop loss, it turns modest winning strategies into losers.

Interested to hear anyone else's experience with this...

CS

I don't claim to be an expert on the matter but here's my thoughts ...

I agree with eightbo who notes some interesting comparisons. It's basically the "casino principle". Casinos make more profit than the probabilities of their games would predict. That's because the casino never stops when losing as players do. So when the returns revert back to the probability the casino recovers it's losses ... they play the ultimate loss recovery strategy! You would never open a position with a blind bet so why close it with one which is effectively what a stop-loss is (other than that the prices has shown a few ticks trend). A stop-loss puts you in the same position as the guy on the roulette table who walks away when he has lost his £100 limit. The stop-loss ensures that if the prices move against you you will always trade out at a loss and never give the price the chance to recover. Whether that is a good or bad strategy is down to a simple equation ... (Losses Stopped Out) - (Profits/Losses if Position Held). Time to dust off the Practice Mode.

[jamesg46]

Loving the new screen setup

[Dallas]

Looking forward to dipping in and out of Cheltenham today between the golf

[Derek27]

Mucking Brilliant Paddy Power H'cap Chase.

[Emmson]

Very sweet when you have most green on the winner, I am also surprised at how much was traded on that race.