Hi All,
over the last few years I have developed a very good machine learning model to predict the outcome of horse races. Using a few simple rules for dutching it has over 50% strike rate over all UK races in the last 18 months. Problem is my wife hates gambling, so while I have used it myself I have mostly paper traded in the hope that I could convince her that it works.
Anyway I thought ok so if I trade, it's not the same as 'betting' and she seems warmer to this idea. I am totally new to trading, my mind has been firmly fixed on picking a winner.
My question is does anyone use dutching as a trading strategy and if not do you think it's possible to turn what I have into one?
Betting to Trading
I think it may be possible for it to be profitable providing you've managed to test it on a large enough sample over the 18 months. I've been doing something similar for the last 10 years. In my case I use various indicators to select 5-6 runners against the field with a minimum/maximum field size. Typically the selections may be 3rd/ 4th in the betting order with 2 or 3 outsiders which I believe could outrun their odds.
In fairness I never recorded the results but I've been doing it successfully for so long that I'm confident in the method.
You may want to look at the Bet Angel dutching videos on youtube. You can certainly Dutch the runners at the time of the race using the Dutching tool. In my case I wanted to automate it by nominating the runners in advance. However in the automation file you can only nominate 2 runners by name unless you open up more instances of Bet Angel. I did ask if there was any way to nominate 4/5 runners by name using other means but no one could answer that one. I've therefore tended to do it manually.
If you want to go ahead with it just test it using £2 stakes.
Good luck
In fairness I never recorded the results but I've been doing it successfully for so long that I'm confident in the method.
You may want to look at the Bet Angel dutching videos on youtube. You can certainly Dutch the runners at the time of the race using the Dutching tool. In my case I wanted to automate it by nominating the runners in advance. However in the automation file you can only nominate 2 runners by name unless you open up more instances of Bet Angel. I did ask if there was any way to nominate 4/5 runners by name using other means but no one could answer that one. I've therefore tended to do it manually.
If you want to go ahead with it just test it using £2 stakes.
Good luck
Ah well I want to do it full time and she won't let me do 'gambling' full time because that would mean I have a gambling problem. I've always wanted to do trading anyway but I like to build models and the data was readily available to build for betting. It seems harder to get hold of and more messy for trading.
I know women have been fighting for equal rights for a long time, but sometimes it goes way beyond equal, to full control.Xeres wrote: ↑Mon Jun 03, 2019 11:10 pmAh well I want to do it full time and she won't let me do 'gambling' full time because that would mean I have a gambling problem. I've always wanted to do trading anyway but I like to build models and the data was readily available to build for betting. It seems harder to get hold of and more messy for trading.
Shouldn't you have a say? Or maybe my stance is the reason why I'm single.
-
ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
So the issue is she doesn't trust you, and your plan is to mislead her into thinking trading isn't gambling. Good luck with that one.
It might be more constructive to help her to understand that everything from running a shop to the insurance industry is 'gambling'. What about bookmakers or city fund managers, they're gamblers too. The danger is when you undertake anything you do wrecklessly, and that's as much of a danger when crossing the road drunk as it is executing an idea you've spent 18 months working on.
You mention a "50% strike rate", you'll know that on its own that number is meaningless, so you might want to express that in terms of value if you want to help someone to understand how you're not 'gambling'. You're simply buying and selling at good prices and acting as a broker for those who aren't.
It's perfectly normal for people to be concerned about what they don't understand, and education is much more powerful than expecting someone to change their attitudes by using blind faith alone or pulling the wool over their eyes.
It might be more constructive to help her to understand that everything from running a shop to the insurance industry is 'gambling'. What about bookmakers or city fund managers, they're gamblers too. The danger is when you undertake anything you do wrecklessly, and that's as much of a danger when crossing the road drunk as it is executing an idea you've spent 18 months working on.
You mention a "50% strike rate", you'll know that on its own that number is meaningless, so you might want to express that in terms of value if you want to help someone to understand how you're not 'gambling'. You're simply buying and selling at good prices and acting as a broker for those who aren't.
It's perfectly normal for people to be concerned about what they don't understand, and education is much more powerful than expecting someone to change their attitudes by using blind faith alone or pulling the wool over their eyes.
-
ShaunWhite
- Posts: 9731
- Joined: Sat Sep 03, 2016 3:42 am
A successful team is one where everyone is pulling in the same direction. Two people saying f-you I'll do what I want is doomed. The 'blackball' voting system has stood gentleman's clubs in good stead for generations on that principal, and 'all in or neither in' seems sound logic for maximum peace and productivity. This is why choosing a good squad is so important but most people just end up with the first person who flutters their eyelashes at them.
I had a girlfriend in the 1990s (at a time when I wasn't betting) who completely lost her temper with me when I had the audacity to tell her there's a lot of skill involved in betting on horse racing. Her view was that its pure luck. One year later and long after we ended our relationship I could sense she realised her ignorance but she never did admit she was wrong - something that was beyond her capabilities. This thread just reminded me of her. [Sorry, this should be on the rant thread.ShaunWhite wrote: ↑Tue Jun 04, 2019 1:51 amA successful team is one where everyone is pulling in the same direction. Two people saying f-you I'll do what I want is doomed. The 'blackball' voting system has stood gentleman's clubs in good stead for generations on that principal, and 'all in or neither in' seems sound logic for maximum peace and productivity. This is why choosing a good squad is so important but most people just end up with the first person who flutters their eyelashes at them.
]-
johnsheppard
- Posts: 284
- Joined: Mon Feb 04, 2019 6:00 am
- Location: Cairns Australia
Shauns advice is pretty good if you were to ask me.ShaunWhite wrote: ↑Tue Jun 04, 2019 1:27 amIt's perfectly normal for people to be concerned about what they don't understand, and education is much more powerful than expecting someone to change their attitudes by using blind faith alone or pulling the wool over their eyes.
The way to go is to help them understand. If they can't understand (its a possibility) you have a problem because it means one of two things;
You have to pull the wool over their eyes and accept the probably inevitable outcome....or
You have to choose what you'd rather do. Make money the way you choose or have a wife.
That's stupid, you have "the Holy Grail" and want to saw off just a little part of it.Xeres wrote: ↑Mon Jun 03, 2019 8:24 pmHi All,
over the last few years I have developed a very good machine learning model to predict the outcome of horse races. Using a few simple rules for dutching it has over 50% strike rate over all UK races in the last 18 months. Problem is my wife hates gambling, so while I have used it myself I have mostly paper traded in the hope that I could convince her that it works.
Anyway I thought ok so if I trade, it's not the same as 'betting' and she seems warmer to this idea. I am totally new to trading, my mind has been firmly fixed on picking a winner.
My question is does anyone use dutching as a trading strategy and if not do you think it's possible to turn what I have into one?
Women is a simple creature, buy her a diamond ring from your supersystem's earnings and she'll forget everything.
I have a dutching strategy with just one simple rule that has a 100% strike rate on any market you choose - and I'll give it to you.
Dutch every horse in the race.
-
wearthefoxhat
- Posts: 3219
- Joined: Sun Feb 18, 2018 9:55 am
I vaguely remember a spreadsheet (freely available) called the credit crunch killer system. You would attempt to dutch every selection in the race by adding x% to the whole race/odds, dependent upon number of runners, input "new odds" into Betfair and let the market move accordingly.
The ones that didn't match pre-off, could be traded out as a loss. (doesn't mean they'll go against you). It was surprising how many would be matched, guaranteeing a win. Also it was a good way of learning how the market moved.
i think as shaun mentioned earlier, the definition of a 50% strike rate will make or break whether this is a viable strategy. you mention that you opt for higher odds runners, which is good, as this will potentially stear you above evens when managing your selections.
i had a dutching strategy a while back that produced a 70% strike rate. the probelm was, on a combined odds basis, i was only getting about 1.3 odds, thus was failing overall. i then adjusted this to scoop out the lower odds runners and redressed the balance somewhat but have yet to do any serious work on making it a bread 'n butter strategy.
at the end of the day, only you will know what this 50% represents, but you really do need to continually re-assess your records, (in relation to those combined odds) to see where you can penny pinch!
i had a dutching strategy a while back that produced a 70% strike rate. the probelm was, on a combined odds basis, i was only getting about 1.3 odds, thus was failing overall. i then adjusted this to scoop out the lower odds runners and redressed the balance somewhat but have yet to do any serious work on making it a bread 'n butter strategy.
at the end of the day, only you will know what this 50% represents, but you really do need to continually re-assess your records, (in relation to those combined odds) to see where you can penny pinch!
Some great comments 
Yes appreciate strike rate is meaningless without context, the system has generated just over 1,000 points using a modified version of the Kelly criterion placing all bets around 10am and no trading out.
There are times of significant downtrends as you'd expect and that is what my wife latches on to.
In my simple mind I was thinking of a kind of hybrid system but not sure that is a good idea, the emotions might get the better of me
Yes appreciate strike rate is meaningless without context, the system has generated just over 1,000 points using a modified version of the Kelly criterion placing all bets around 10am and no trading out.
There are times of significant downtrends as you'd expect and that is what my wife latches on to.
In my simple mind I was thinking of a kind of hybrid system but not sure that is a good idea, the emotions might get the better of me
Your postion sounds familiar.
Dutching can be a trading strategy. The real question is, can you put theory into practice and show a profit over several months? Strike rate is a consideration but finding value is essential.
Paper losses do not invoke the same reactions that come with real losses. Some losses are within the model, others can be a problem. Understanding or accepting the difference is not easy.
My advice. Allocate part of your entertainment budget, to try actual “trading” with small stakes. Keep complete records, available for scrutiny by your wife. You may find that achieving consistent profit is harder than you expect, or not possible.
If your strategy is successful, then look for refinements and gradually try scaling up to a serious income.
Dutching can be a trading strategy. The real question is, can you put theory into practice and show a profit over several months? Strike rate is a consideration but finding value is essential.
Paper losses do not invoke the same reactions that come with real losses. Some losses are within the model, others can be a problem. Understanding or accepting the difference is not easy.
My advice. Allocate part of your entertainment budget, to try actual “trading” with small stakes. Keep complete records, available for scrutiny by your wife. You may find that achieving consistent profit is harder than you expect, or not possible.
If your strategy is successful, then look for refinements and gradually try scaling up to a serious income.