The thing that interests me is how many decisions a manual trader is actually making.
Does a manual trader make hundreds, or tens, or a few calculations in a given moment?
How many decisions does it take to commit to a pattern that they know will increase the probability of the trade being right?
A bot can sit and wait for a big back order to come in, say at 3.00 and jump in at 2.98 and ride it down past 2.9, having an exit plan on the way. The bot can be watching other markets at the same time. It's adventurous in my experience to get a bot to monitor more than 20 or so different decisions at anyone time on a single market, but I'm sure people are building bots that do more.
Bots V Manual Traders
-
marksmeets302
- Posts: 527
- Joined: Thu Dec 10, 2009 4:37 pm
Most bots will be targeted at specific situations. I doubt there will be many people that build their bots to take care of many market situations.
One strategy per bot, and one bot per account. Makes it very easy to monitor if your stuff is working. Drawback is that your betfair points are scattered over several accounts.
One strategy per bot, and one bot per account. Makes it very easy to monitor if your stuff is working. Drawback is that your betfair points are scattered over several accounts.
marksmeets302 wrote:Most bots will be targeted at specific situations. I doubt there will be many people that build their bots to take care of many market situations.
One strategy per bot, and one bot per account. Makes it very easy to monitor if your stuff is working. Drawback is that your betfair points are scattered over several accounts.
Ah no. On a Saturday I will track around 60 different football markets. So each minute on a 1 second cycle guardian checks each market. I don't have the same strategy for each market. There is no way any manual trader would get close.
Then on the same account at that same time but a different instance of bet angel I am tracking 100 horse racing markets and 100 greyhound markets.
On a Saturday I have five or six instances of bet angel open and running. On a weekday I have at least three open.
Some markets i'm always watching in guardian, cycling say five markets on a 1 second basis.
It's just impossible for a manual trader to track everything guardian tracks for me. Hats off to Peter and the Bet Angel team for providing this functionality.
Last edited by gazuty on Tue Apr 28, 2015 1:25 pm, edited 1 time in total.
-
NeilSpence
- Posts: 78
- Joined: Sun Nov 17, 2013 11:58 am
Gazuty,
Are all the strategies you are using your bots for profitable?
Are all the strategies you are using your bots for profitable?
Yes, other than harness racing which is break even (it generates commission and reduces pc and my future pc liability).NeilSpence wrote:Gazuty,
Are all the strategies you are using your bots for profitable?
See also post below. Some automations or ideas that I have from time to time turn out to be losers. I stop running them or tweak them etc.
But right at the moment I'm in a phase where there is nothing in development as such.
Last edited by gazuty on Wed Apr 29, 2015 2:54 am, edited 1 time in total.
I should be clear that not every idea has developed into a great bot. And I've developed any number of losing bots.
My most successful automation was developed after two years of Saturday afternoon manual trading and market observation. I then have consistently refined and tweaked. This has delivered the money to experiment. Without this discovery I might have given up long ago.
Other automations have come through having an idea, building the rules, deploying with small stakes and seeing if profitable.
My ideas are very simple and are generally based around expected return and harvesting over the long run. So I'll use low stakes and run something for months before I decide if it's a long term winner or long term loser.
Sometimes my ideas are just plain wrong and turn out to be a guaranteed loser after I've given them more thought.
Finally, I keep track of what works and what doesn't.
Hope that helps.
My most successful automation was developed after two years of Saturday afternoon manual trading and market observation. I then have consistently refined and tweaked. This has delivered the money to experiment. Without this discovery I might have given up long ago.
Other automations have come through having an idea, building the rules, deploying with small stakes and seeing if profitable.
My ideas are very simple and are generally based around expected return and harvesting over the long run. So I'll use low stakes and run something for months before I decide if it's a long term winner or long term loser.
Sometimes my ideas are just plain wrong and turn out to be a guaranteed loser after I've given them more thought.
Finally, I keep track of what works and what doesn't.
Hope that helps.
Hi, Gazuty,
You say you moved your manual trading into making a bot, the postion i'm in is that i can make small amounts of money per market mostly between £1 to £3 and the odd one higher with the odd small loss, you said that you would probably have giving up if you didn't get your bots going was this because you were in the same postion?
What i'm saying is if you can make a average profit of £2 per market manual trading the amount of money you make isn't worth the time spent trading, but if you can use a bot it is?
You say you moved your manual trading into making a bot, the postion i'm in is that i can make small amounts of money per market mostly between £1 to £3 and the odd one higher with the odd small loss, you said that you would probably have giving up if you didn't get your bots going was this because you were in the same postion?
What i'm saying is if you can make a average profit of £2 per market manual trading the amount of money you make isn't worth the time spent trading, but if you can use a bot it is?
Hi Gazuty, I'm interested to know whether the majority of your bots are rule-based automation, or do you use Excel extensively? I was initially attracted to BetAngel as I had a strategy that required complex calculation in real time and BetAngel allowed me to automate that. I realise, somewhat belatedly, that I've totally ignored non Excel automation and that I may be denying myself opportunities that exist there.gazuty wrote:
On a Saturday I have five or six instances of bet angel open and running. On a weekday I have at least three open.
Some markets i'm always watching in guardian, cycling say five markets on a 1 second basis.
It's just impossible for a manual trader to track everything guardian tracks for me. Hats off to Peter and the Bet Angel team for providing this functionality.
I don't use excel any more, I only use guardian with rules based automation. IMHO guardian automation is very powerful. Some of my automations have over 100 rules each with multiple conditions and guardian works fine.Wyndon wrote:Hi Gazuty, I'm interested to know whether the majority of your bots are rule-based automation, or do you use Excel extensively? I was initially attracted to BetAngel as I had a strategy that required complex calculation in real time and BetAngel allowed me to automate that. I realise, somewhat belatedly, that I've totally ignored non Excel automation and that I may be denying myself opportunities that exist there.