Frosty's trading bot goes live part 2
- Thread starter frostengine
- Start date
Is that a typo?
Yes typo, sorry about that
I'm really curious where the bot can actually profit and not a couple hundred bucks either. It doesn't really profit on huge moves to the upside and not so great on choppy days either.... Frost, you have yet to see ONE day where the bot can make you SUBSTANTIAL gains. It's very good at chipping away and then BAM, big loss(es) ensue.
Needs to be the other way around in my opinion.
Needs to be the other way around in my opinion.
frost,
out of how many of the past 18 live days would you say that your bot has gone +200$ into the profit. iv been playing around with the idea of daily profit targets which would help smooth the equity curve, prevent days like this, help guarantee a positive return while giving up potential profit. have you done any backtests where you have the bot stop for the day after it hit a 200$ profit target. id be curious to hear how that effects overall profits, %profitable days, drawdowns and w/l ratios.
out of how many of the past 18 live days would you say that your bot has gone +200$ into the profit. iv been playing around with the idea of daily profit targets which would help smooth the equity curve, prevent days like this, help guarantee a positive return while giving up potential profit. have you done any backtests where you have the bot stop for the day after it hit a 200$ profit target. id be curious to hear how that effects overall profits, %profitable days, drawdowns and w/l ratios.
sometimes consistency is rightly exchanged for profit.
Dear Frost,
I retired from posting in this forum, but will set that aside for one note. I absolutely cannot bear to see you struggling as such in real time bleeding real money this way. Having extensive mechanical system (bot) writing experience in general and ER in particular some years ago, let me share a couple of things with you:
#1: Forget whatever you saw in historical backtesting. That's a fallacy. Forward live-trade results are upside down. That is reality. Your bot configuration is not even remotely near robust enough to run profitably in real time, especially on an intraday basis.
Long story short, what you see right now is what you'll continue to get. Skip the details and trust me on that fact. Or you will inevitably trust the result of a hemorrhaged account in the end.
A system shakedown thru rina software (www.rinafinancial.com) with walk-forward scenarios and monte carlo testing would have all but assuredly proven this bot unfit. Then again, the software itself costs pretty much what you are in the hole right now so that's a wash.
I used this software for years myself, and learned one helluva lot about bots that look great in backtest versus reality when wrung thru the ringer of mathematical scenarios.
#2: Forget about writing an intraday trading bot, period. That is beyond the scope of <b>most</b> retail traders to perfect. The profit/loss edge is too thin when confined to intraday price carriage only. You have hope of writing a profitable bot via swing trade fashion, i.e. overnight holds.
Writing an intraday bot will probably cost you $10,000 ~ $20,000 lost capital in testing and months if not years of time to perfect. It is the thinnest edge for trading success, therefore the toughest timeframe to succeed in. Limiting potential profit size on the big wins to intraday range while keeping max loss fixed is an albatross to bots.
Please don't let anyone say you should fear holding open trade-bot positions overnight due to unexpected loss. Disaster scenarios can be overcome with account = position size. You are already enduring disaster-type loss in reality as it is... overnight holds will not be any worse than what you are accomplishing in real time with real money right now!
#3: Bot writing = trading is equally hard mentally & emotionally as discretionary trading. I've done plenty of both, so have many others here. The fantasy of letting the computer print money in automated fashion while we remain detached from the process is pure folly. I think you have experienced quite a bit of that fact by now, and others have watched the ride in sympathy as well.
I no longer bot trade and never will again myself, but it's one choice trader's have to operate in this profession. Now you realize bot trading is every bit tough in every way AND costly education in real $$ as discretionary trading is.
==
Please feel free to email me with questions or concerns and I'll discuss bot writing = testing in detail with you. I'm not offering trade advice either way, but I will state from my experience with bot systems that yours is upside down. Small profits and large losses (comparative) is all it has ever produced. Regardless of how, why or what excuses, that is the real dollar summation. If you shake down this specific system thru monte carlo walk-forward scenarios, I would expect the bell curve of probability would be frightening. Or sobering.
Take great care, and good luck
I retired from posting in this forum, but will set that aside for one note. I absolutely cannot bear to see you struggling as such in real time bleeding real money this way. Having extensive mechanical system (bot) writing experience in general and ER in particular some years ago, let me share a couple of things with you:
#1: Forget whatever you saw in historical backtesting. That's a fallacy. Forward live-trade results are upside down. That is reality. Your bot configuration is not even remotely near robust enough to run profitably in real time, especially on an intraday basis.
Long story short, what you see right now is what you'll continue to get. Skip the details and trust me on that fact. Or you will inevitably trust the result of a hemorrhaged account in the end.
A system shakedown thru rina software (www.rinafinancial.com) with walk-forward scenarios and monte carlo testing would have all but assuredly proven this bot unfit. Then again, the software itself costs pretty much what you are in the hole right now so that's a wash.
I used this software for years myself, and learned one helluva lot about bots that look great in backtest versus reality when wrung thru the ringer of mathematical scenarios.
#2: Forget about writing an intraday trading bot, period. That is beyond the scope of <b>most</b> retail traders to perfect. The profit/loss edge is too thin when confined to intraday price carriage only. You have hope of writing a profitable bot via swing trade fashion, i.e. overnight holds.
Writing an intraday bot will probably cost you $10,000 ~ $20,000 lost capital in testing and months if not years of time to perfect. It is the thinnest edge for trading success, therefore the toughest timeframe to succeed in. Limiting potential profit size on the big wins to intraday range while keeping max loss fixed is an albatross to bots.
Please don't let anyone say you should fear holding open trade-bot positions overnight due to unexpected loss. Disaster scenarios can be overcome with account = position size. You are already enduring disaster-type loss in reality as it is... overnight holds will not be any worse than what you are accomplishing in real time with real money right now!
#3: Bot writing = trading is equally hard mentally & emotionally as discretionary trading. I've done plenty of both, so have many others here. The fantasy of letting the computer print money in automated fashion while we remain detached from the process is pure folly. I think you have experienced quite a bit of that fact by now, and others have watched the ride in sympathy as well.
I no longer bot trade and never will again myself, but it's one choice trader's have to operate in this profession. Now you realize bot trading is every bit tough in every way AND costly education in real $$ as discretionary trading is.
==
Please feel free to email me with questions or concerns and I'll discuss bot writing = testing in detail with you. I'm not offering trade advice either way, but I will state from my experience with bot systems that yours is upside down. Small profits and large losses (comparative) is all it has ever produced. Regardless of how, why or what excuses, that is the real dollar summation. If you shake down this specific system thru monte carlo walk-forward scenarios, I would expect the bell curve of probability would be frightening. Or sobering.
Take great care, and good luck