How long does it take to judge whether a system is working?

Trading is often about how to take the appropriate risk without exposing yourself to very human flaws.
Post Reply
FrogThimble
Posts: 124
Joined: Mon Dec 11, 2017 6:26 pm

Dallas wrote:
Fri Apr 27, 2018 9:07 pm



Flat season began about 2-3 weeks ago and NH is now ending, there is the odd jumps meeting still on throughout the summer but its basically over till September when the flat winds down and the jumps begin again
Thanks for the information, Dallas. It's not what I'd hoped to hear but at least I know. Trust me to device a strategy based around long races just as they're about to end... :roll: Oh well... I'll give the shorts ones a fair try seeing as they're the only game in town. :D
User avatar
Dallas
Posts: 22642
Joined: Sun Aug 09, 2015 10:57 pm
Location: Working From Home

FrogThimble wrote:
Fri Apr 27, 2018 9:13 pm
Dallas wrote:
Fri Apr 27, 2018 9:07 pm



Flat season began about 2-3 weeks ago and NH is now ending, there is the odd jumps meeting still on throughout the summer but its basically over till September when the flat winds down and the jumps begin again
Thanks for the information, Dallas. It's not what I'd hoped to hear but at least I know. Trust me to device a strategy based around long races just as they're about to end... :roll: Oh well... I'll give the shorts ones a fair try seeing as they're the only game in town. :D
If you do have to park the strategy for a few months during summer at least you'll have something to pick up again in September, as the chart shows once it starts up again it doesn't take long to get back in the full swing.
User avatar
ShaunWhite
Posts: 9731
Joined: Sat Sep 03, 2016 3:42 am

Flat races aren't all sprints. You might also find that the going is a consideration too if distance is.

It's certainly worth you learning a bit about race types and grades to supplement your analysis.

Also if the strategy depends on how the favourite runs you might want to look at a service like pacecards to quickly eliminate some likely to spoil your day.
weemac
Posts: 1216
Joined: Mon Sep 16, 2013 8:16 pm

Totally different ball game, imo.
FrogThimble
Posts: 124
Joined: Mon Dec 11, 2017 6:26 pm

I'll see how it pans out next week with my existing conditions (other than distance). If it goes wrong at least I'll be picking up clues as to why as I go along.
FrogThimble
Posts: 124
Joined: Mon Dec 11, 2017 6:26 pm

ShaunWhite wrote:
Fri Apr 27, 2018 9:37 pm
Flat races aren't all sprints. You might also find that the going is a consideration too if distance is.
I'm curious... what is the commonly accepted definition of a sprint race in terms of ranges of distance?
User avatar
ShaunWhite
Posts: 9731
Joined: Sat Sep 03, 2016 3:42 am

FrogThimble wrote:
Fri Apr 27, 2018 10:04 pm
ShaunWhite wrote:
Fri Apr 27, 2018 9:37 pm
Flat races aren't all sprints. You might also find that the going is a consideration too if distance is.
I'm curious... what is the commonly accepted definition of a sprint race in terms of ranges of distance?
5 or 6f
FrogThimble
Posts: 124
Joined: Mon Dec 11, 2017 6:26 pm

ShaunWhite wrote:
Sat Apr 28, 2018 12:47 am

5 or 6f
Thanks Shaun. :) They look a lot more fun for the pre-off people (which isn't my thing) than for the in-play people... but I'll see how it goes.
User avatar
ShaunWhite
Posts: 9731
Joined: Sat Sep 03, 2016 3:42 am

FrogThimble wrote:
Sat Apr 28, 2018 8:57 pm

They look a lot more fun for the pre-off people (which isn't my thing) than for the in-play people... but I'll see how it goes.
Before someone else says it... "Every market has it's strategy and every strategy has it's market".

If you don't like the sprints there will be plenty of middle distance and stayers races on the flat.

Looking back I see that you're using a fairly basic back to lay method, I'd have thought that's more suited to the flat than the NH actually especially with you being interested in the longer slower moving markets; you don't have the same risk of poor jumping or fallers making it so hard to get out at the target red.

It might also be worth you looking at the breakdown between non handicaps and handicaps which are of course designed to produce closer races.

You said you don't use historical data, but you only use maths and know nothing about racing, so what is it that you're doing? Setting a figure for a win and one for a loss and hoping for a certain strike rate, or something much deeper?
FrogThimble
Posts: 124
Joined: Mon Dec 11, 2017 6:26 pm

ShaunWhite wrote:
Sun Apr 29, 2018 1:28 am
FrogThimble wrote:
Sat Apr 28, 2018 8:57 pm



You said you don't use historical data, but you only use maths and know nothing about racing, so what is it that you're doing? Setting a figure for a win and one for a loss and hoping for a certain strike rate, or something much deeper?
If the race meets my conditions I bet on the Favourite to win just before off time using By Liability (back and lay) stakes . Then, during the race, I try to trade out for about between 5% to 7% of what I'd get if the horse were to win. I have two possible escape points if the movement doesn't go in my direction. I trade for a Red if the market moves against so that I'd lose about 3 times the amount I'd win on a Green trade or if my horse ceases to be Favourite... whichever happens first (in most cases the latter is more likely to happen than the former unless my horse has a very short price to start with). Which of those two is the most costly depends on the relative prices of the horses - and it's the price relationship between the Favourite and the next in the field before off which dictates whether I play anyway. So I'm trying to avoid races where there might be a statistcally likey very early crossover in price.
FrogThimble
Posts: 124
Joined: Mon Dec 11, 2017 6:26 pm

Day 7:

Today 8 races met the conditions of my strategy taking my total trades up to a nice round 100. Although I made a decent profit today I feel it was more by luck than by judgement. 6 of the the races played out "normally" allowing me to get out for small profits or small losses but I also had 2 races where all hell broke loose really and I was lucky to come out on the upside of those overall.
page 7.jpg
In the Wetherby 4:25pm I was trying to trade out for a small loss at exactly the moment that the market swung wildly in my favour and instead of being down 19p I ended up gaining £1.41...

Then in the Wetherby 5:30pm I had pretty much the reverse scenario. I was greening up to lock in a 5p profit at precisely the moment that the market swung wildly against me and went 48p down instead.

These two markets were my biggest gain and biggest loss so far in percentage terms so it confirms to me how volatile these short races are.

I was also unable to exert any control over my position in the Wetherby 6:00pm but ended up OK on that one in the end.

Anyway, as a consequence my graph looks a bit crazy tonight.

Day 7.jpg

Today's strike rate: 75% (6/8)
Overall strike rate: 72% (72/100)
Today's profit: 8.6% growth of my bank (£12.97 to £14.08)
Overall profit: 40.8% growth of my bank over 7 trading days (£10.00 to £14.08)
You do not have the required permissions to view the files attached to this post.
User avatar
ShaunWhite
Posts: 9731
Joined: Sat Sep 03, 2016 3:42 am

So the strategy is to back to lay clear favourites. I wish you well but it's going to be really difficult to do this consistently especially as horses running styles become more exposed. You will start to find that early leaders don't shorten as much as you hope because the inplay specialists won't see this as an indication the horse will win any more often than sp suggests.

With you being new to racing you might not appreciate that horses don't just run as fast as they can from start to finish. People who make money from your type of strategy usual only manage to do so by factoring in the horses running style.

I really do think you should try looking at the historical data, you can leave your spreadsheet running overnight as wake up to every permutation of profits and stops for every surface and distance, for every variation of favorite prices over the last 100000 markets. It will save an aweful lot of guesswork. Although, if you do you'll join a group of about 1000 people who have tried and failed to find a simple edge in this area.

With all due respect I don't see any application of maths, it's just picking random profits and stops. You will be very very lucky to have found a magic combination and the pl from your last 60 races (ie almost zero) shows its not quite as easy you might have hoped. I'd never deter someone from trying but I'd use your random strategy as an opportunity to learn about the fundamentals.

The markets are brutally efficient, I'd guess that after 500 races you'll find that you are in loss by approximately the amount of commission you've paid.
FrogThimble
Posts: 124
Joined: Mon Dec 11, 2017 6:26 pm

ShaunWhite wrote:
Sun Apr 29, 2018 10:37 pm


With all due respect I don't see any application of maths, it's just picking random profits and stops. You will be very very lucky to have found a magic combination and the pl from your last 60 races (ie almost zero) shows its not quite as easy you might have hoped. I'd never deter someone from trying but I'd use your random strategy as an opportunity to learn about the fundamentals.

The markets are brutally efficient, I'd guess that after 500 races you'll find that you are in loss by approximately the amount of commission you've paid.
The stops are not exactly random. I have two Servants that I've set up to take me out at very specific points. I was finding that they were working reasonably well on the longer races. However, on the shorter races I find it harder to sit back and let them get on with it due to the volatility of the prices and this causes me to intervene to try to get any scrap of profit that I can before it goes red - hence on the shorter races I'm winning less on my winners but losing the same on my losers as I would on the longer races.

So I think it's fair for you to say that my profits and stops on the Flat races are random but on the Jump races I'd argue that they are not because I have been letting the Servants deal with the Jump races unaided (or unhindered) by human intervention.

You also make a very fair point about the poor performance of my system over the last 60 races. The best part of the graph is without doubt the first 20 races and the most interesting thing about that section of the graph is that almost all of those races were jump races.

Races 1-20: 17 Jumps, 3 Flats
Races 21-40: 14 Jumps, 6 Flats
Races 41-60: 13 Jumps, 7 Flats
Races 61-80: 11 Jumps, 9 Flats
Races 81-100: 13 Jumps, 7 Flats

I have a 78% strike rate on the Jump races and just a 58% strike rate on the Flat races (55% on the 5-6 furlong ones). A strike rate of roughly 66% on this is break-even, I think.

All of the races in the next 20 will be Flat races and I very much expect to take a hit because of that based on the limited data I've gathered so far.

I never thought it would be easy though. If I thought it would be easy I'd be playing these trades with a far bigger bank than I am doing - I'd have kicked off with £100 or £200 in my bank if I were sure of my theories. I think the small bank I am using is a reasonable reflection of my confidence (or lack thereof) in my system. ;)
User avatar
ShaunWhite
Posts: 9731
Joined: Sat Sep 03, 2016 3:42 am

What I meant by random is that the points you close at weren't selected with anything specific in mind. Such as the average inplay minimum a clear favorite in the pre race betting reaches. 7% just felt doable more often than not? I wasn't suggesting you just closed your eyes and clicked :)

More sophisticated automation might help as you could include logic to let your profits run further, or to reduce your loss with a stop that moves intelligently, maybe nudging down in steps as different profit levels have been reached. It would be ideal if the clear leaders that just head off into the sunset paid really well to help offset the occasional one who never gets going and screws you over. Remember that if the average price you backed at was 3.0 then probably 33% of them were winners. The difference between a winning strategy and a losing one is often just a few percent.

It takes 300ms to blink and 200ms to react to something and click. In that time, BA could have processed your logic rules 25 times with a dozen BF price changes. Unless you have a talent for reading how horses are looking in running, full automation really is the only place to be for inplay horses. You need all the speed you can get when you're fighting loads of other people to grab that tenner on the ladder.

You're wise to keep it small and the best thing of all is that you're enjoying it. Sorry if I came across as cynical but my intention is just to try and give you a steer, if not to profits then definitely away from losses. Your strategy is a well known one, no need for everyone who tries it to pay the full cost of learning the pitfalls.

With your strategy being quite long duration back to lays I still think you'll do better on the flat. Flat sprints won't be ideal though as how they break from the stalls can be a do or die within seconds.
FrogThimble
Posts: 124
Joined: Mon Dec 11, 2017 6:26 pm

I'm fascinated by your suggestion that I might do better on the Flats, Shaun. I hope you're right about this as there's precious little else on offer right now. I think I am struggling with the more rapid rhythm of these markets though and it does indeed seem to be do or die on the short races.

I very grateful to you for defining sprint races for me as 5 to 6 furlongs as they are my worst results so far. From today I'm not playing them and just going with 7 furlongs and higher so long as the prices qualify for my system.

The only real hint of sophistication that I have in my Servants is that the price differences between Fav and 2nd Fav that are acceptable to me are different depending upon the price of the Fav... By that I mean I'm not looking for the same price gaps between a short price Fav and its 2nd Fav as I am between a longer priced Fav and its 2nd Fav.

I have 4 different price bands for Favourites and each of those 4 bands demands a different gap between the Fav and 2nd Fav.

Anyway, I appreciate your input and I didn't take your comments as being cynical. :)
Post Reply

Return to “Trading Psychology”