This is just some random thoughts. Risk management, two different scenarios: 1. One trading approach I found seldom used except by someone like Taleb is an approach used by venture capitalists: To get payoffs from high payoff but low probability ventures, you need to invest (e.g., long DOTM options) in lots of them. A 1-10% wins in those cases will have handsome payoff. 2. The opposite for high probability of win cases, with steady low payoff but a few times with huge losses (e.g., write naked OTM calls and puts): You leverage like hell for a few trades, then take your money and run - stop trading. Think Karen the super trader stopped after she made $10-$20 M laughing all the way to the bank instead of kept going at $50 M, $100 M, then went broke.
No, don't stop. Take out a lot of cash, and trade with a smaller size. But continue till it does not work anymore. Why would you stop trading a very profitable system? The real idiots would continue to compound till eternity. There are quite a few of them on ET.
I will say I been most often in my life WRONG when it comes to trading long term, but when I been wrong, I have "bet" one unit, and when I have been RIGHT, I have bet 50 units for long term trading. Doesn't sound correct of how I can trade this way considering I trade same amount of contracts on each attempt but when I check my stats, end results are this way when I hedge. I average being wrong 8 trades, so overnight margins of say on average $3,000 and options $3,000, so expensive way to trade, so after 8 trades and fees and some losses, it adds up, so last trade pays for the mistakes of not being extreme enough, taken me 24 tries when crude oil went to 147 and I lost count of 7 years of Eurodollars, more than 50 I am sure, had to redesign for that market, doubt I live long enough for that to happen again. So in a strange way, it can be thought of am doing #1, pay offs fairly large, but sporadic and over the time wise of year(s), so a win would have to be completion of 75% range. Require huge patience. I can see and would be testing #2 at some point, I think it can be done easier than most think, and if you understand "Swing" lengths of longer term price action, you be able to do at least half of the long term swing at very least. Many times at Gann's 50% area has a turn around for so far before completing the cycle. I don't think it viable to do when it becomes 50/50, but I think it better to sell slightly OTM options than deep and make it a spread of buying 90% away, so you can't lose the farm. Absolutely, you have something that works, don't even think of doing less, it anything do 50% more, take some insurance to sponge off that risk.
This arises for quite a lot of technical reasons, Luke: for a start, you need ultra-fast reactions and execution of a kind which it would really be terribly difficult to achieve manually; specialist facilities, software, and so on, to have any real, long-term chance of doing it profitably. Be aware, though, that sometimes in trading forums, members use the word "scalping" when they actually mean something that isn't technically scalping at all ... something more like "fast trading from relatively fast-moving charts, aiming to trade relatively frequently with high-probability entries and take only small numbers of ticks/pips in profit before closing the entire position" - and that's not really the same thing as "scalping", per se, at all: it's just "fast intraday trading". (It's also pretty close to most of what I do for a living, myself - but in no way am I a "scalper"!).
Would you agree in a definition of scalping being in and out of a position for a few ticks, the trade normally going from seconds to some few minutes?
It's not what I mean by "scalping", nor what any professional trader I know myself, means. But I know very well that in some forums (I think actually not in this one, thankfully) it passes as loose vernacular parlance for "scalping". That's to say, the shared understanding of the meaning of "scalping" is vague enough, in some circles, to include both that kind of trading and actual scalping as well!). To me, a trade that's open for "some few minutes" - as most of mine are - isn't even close to scalping: it's a whole different world, and significantly so, given the skills, attributes and equipment needed for scalping. (My colleagues would laugh at me - and rightly - if I claimed to be "scalping"!).
Scalping is not time-based and defined by the amount of expected profit, usually within the 2-3 tick-range.
Well in order to identify a model you have to set the parameters that define that model first, so you have to come to some sort of conclusion of what scalping is or what it is not. Scalping does not refer only to the time frame but also on making small profits (sometimes on huge volume) while the price is changing. The fact that the time frame is short its because of the relative speed the market is moving (the faster the market, the shorter the time frame to get those small profits). We can all agree that shorter than seconds would be HFT domain. So if you are not competing in HFT domain, why would you need to automate in order to have a chance in the first place to survive as a scalper? And if you are competing against HFT, what makes you think that automating yourself as a retail trader from your home would translate into having any chance to beating algos that trade as close to the exchanges as possible, with really expensive high end hardware?
"When I use a word," Humpty Dumpty said, in rather a scornful tone, "it means just what I choose it to mean- neither more nor less." "The question is," said Alice, "whether you can make words mean so many different things." Call me old-fashioned; call me pedantic if you want to ... but "scalping" means "crossing the spread". (Which was doubtless why Mark wisely mentioned that you effectively have to be automated in order to scalp, which is what you asked, and the question I tried to help you by answering. With apologies, I don't have time for "the full 30-minute argument" about it! )