Algorithm and Portfolio Stats - 04/01/2024 - 04/05/2024
Our algorithm is on another successful week. In total, it’s up 0.48% since we started this version in mid-March. This calculation assumes 16.67% allocations into each position, since our system has traded an average of 6 tickers per day during this period. 0.48% isn’t much, but again, that estimate makes some fairly conservative assumptions of our users’ allocation strategies. Our system doesn’t recommend allocations per position, since we want users to do their own due diligence. As such, when calculating overall returns, we’re assuming that allocations are made equally into all positions, and done in such a way that no margin is ever needed. So, if you aren’t leaving most of your capital idle during the day, or you have some experience trading and are able to pick out stronger positions, you’re likely to out-perform this estimate.
I also want to discuss market beta for a moment, to give these results some more context. Given that SPY is down about 1% this week, positive returns here might look even more impressive. But this system is designed to have zero market beta. This is something we’re able to achieve by taking both long and short positions on every individual ticker it’s able to trade. This week, this very much worked in our favor. Since our system can profit off its tickers regardless of which direction they’re moving, red weeks like this don’t have a substantially negative effect on it. This week, this was a great thing for us and helped our system to beat the S&P by a large margin. But at the risk of over-selling, I want to emphasize that there’s a downside to the zero-beta strategy as well. There are going to be green weeks where it under-performs for the same reason. We’ve chosen a zero-beta strategy because the goal of the day trade bot is not to consistently beat the S&P 500, but rather to provide consistent positive returns.
The other thing that I want to note about our results so far is the distribution of returns. In backtests, we’re looking at about a 33% profit rate on trades from this system. That’s not a bad thing, because our winning trades make significantly more than our losing trades lose. As a result of this, we’re expecting many days with this system to be slight losses overall, with occasional large success days that turn this system into a winner. We also see this with individual trades - with a relatively small number of big successes which have a big impact on our bottom line. I’d like to emphasize that this is a feature - not a bug.
I’m fairly happy with cumulative returns during this testing period, but at the same time, we’re always working to improve things. We currently have 1 separate, more experimental algorithm being run on a staff-only test server. This is a number I’m hoping to increase in the next few weeks. This isn’t something we’re doing due to lack of faith in this algorithm, but because we’re always looking for ways to improve it. It’s doing well, but we know it can be better.
Now then, let’s examine some of our big winners and losers from this week. We took 52 trades this week (23 long, 29 short) across 31 unique tickers.
Our most profitable trade of the week was this short on BLDR on Thursday. There were no re-entires here; one position made us 1.43%. I’ve analyzed this position previously, and I stand by the conclusions reached previously: the entry conditions here are fantastic. Momentum is a little sideways before we get in, but it picks up almost immediately after our entry, which made this trade a really easy one to stick with until the end.
Our best long trade of the week was on OXY on Friday. Like BLDR, this was one position with no re-entries, for a return of 1.28%. Most of what I said about BLDR also applies to this trade, but if anything, I think our entry conditions here are stronger. Unlike the BLDR trade, where we entered on mostly sideways momentum, OXY had strong bullish momentum from the beginning. The TK-cross is very strong here, MACD is solid - this is really what we want to see from all of our entries. I would have taken this every time.
Our worst trade of the week was a short on NRG on Tuesday. Over 3 entries, we lost 0.65% here. The real problem with this trade is that momentum is sideways. That’s not a trade-killer in and of itself (our BLDR trade showed this same red flag prior to our entry), but the real problem is that the momentum never improves during the life of the trade. Unlike BLDR, where momentum picked up almost immediately, here it was always lacking in direction. Consider our initial entry, and 2 re-entries. The MACD is okay when we first enter, but by the time we re-enter, it isn’t any stronger - in fact it’s worse. This is a red flag we encourage users to look out for. I probably would have taken that initial entry, but I probably would have refused to re-enter the position, unless MACD had improved significantly.
Now, let’s examine some stats on our portfolio:
Our portfolio had a very strong week - coming in at 0.07% profit, versus SPY’s loss of 1.03%.
Despite being the plurality of our portfolio, we mostly broke even on tech. Our big win was on Communications - with Energy and Industrials further shoring it up. It goes to show: even if your investment thesis focuses on one or two sectors, it’s a good move to diversify.
This coming week, our portfolio is looking similar, but we’re continuing to expand its exposures. Rather than exclusively over-allocating technology as we did some weeks ago, we’re now distributing our excess betas much more evenly. Only Consumer Discretionary, Consumer Staples, Technology, and Utilities heavily differ from the S&P’s exposures.
Our allocations going forward are below. For brevity, all tickers below 0.5% allocation are excluded.
That’s all I have for you tonight. Thank you for reading - and happy trading!
