We have tested FAPturbo back to ancient Egyptian times!
OK…maybe NOT quite that far back! However my friend Dave has been running a series of backtests on FAPturbo and has produced these results:
I’ve started to backtest the strategies on all pairs, beginning with just a 1-month to 3-month period initially, as each test can take 20 to 30 minutes on my PC depending on the level of optimization included. Most of my testing has been on the scalper strategy as it has much fewer input parameters and works on a higher timeframe (15 minutes as opposed to 1 minute for the FapTurbo strategy) so the tests take less time to run.
However, even with this limited range of tests, some interesting results have been achieved.
1. Scalper strategy
Initial findings: using Scalper_MM set to true (for FAPturbo’s own money management) delivers better profits. For example, on the EURGBP pair, it delivered roughly twice as much profit as when MM is set to false, with all other parameters the same.
Changing the value of risk reductor only has an effect when MM is set to false (presumably because MM uses its own settings to control risk in each trade, so it ignores this parameter).
Increasing the lot size does what you’d expect: effectively multiples your profit (or loss) by the lot size used, so using lots at 0.5 will clock up 5 times the profit/loss values obtained at lot size 0.1. Drawdown is usually larger with a larger lot size, though not necessarily by the same multiple. For example, the 0.5 lot size delivered 5 times the profit on EURGBP for a drawdown that was approx 40% higher than that of the 0.1 lot size.
By the way, drawdown level is an important point to bear in mind. Although each of the pairs covered by the scalper strategy delivered profits over my test period, they all suffered drawdowns to varying degrees. I haven’t checked to see what the total would have been across all 4 pairs at any given time, but suffice it to say, that there is a substantial risk involved in consistently trading all 4 pairs at the same time. There’s no rule that says you won’t hit a losing streak of a few trades on all 4 at the same time, and, unless you have good money management in place, your account could be wiped out before the eventual winning trades come through and return you to profit.
Some parameters such as the MA filters had the effect of filtering out a number trades, but on the MA values I’ve tested so far, the filters did not deliver a higher percentage of winners, so the overall profit declined more or less in line with the reduction in the number of trades.
2. FapTurbo strategy
The FT strategy on EURUSD has not worked well during the past month (November 5th to December 5th) delivering losses (depending on settings) of between 1% and 13%.
Over a 3 month period, in the two tests I ran, it delivered a profit of 4% in one case and 230% in the other case. The difference in settings? In the first case, the number of lots was 0. In the second it was set to 0.5. All other parameters were the same (mainly default values). Of course, this is (so far) a very narrow sample of tests, so there are significant risks in drawing any broad conclusions from it. In particular, I think it’s fair to say that any trend-following strategy would have done well on the EURUSD pair between August and November this year, simply by focusing on selling short during that fall of almost 4000 pips. Now that this pair is consolidating, it’s much more difficult to find large winners.
Other points:
The FT team recommend not trading on Fridays, so I ran a few tests with the relevant parameter set to true and then to false to check the effects. The Scalper strategy seemed to cope OK with Friday trades, and in the cases I’ve run so far, the strategy ended up making a few additional trades compared to the non-Friday version, and on balance finished with a slightly higher profit.
The FT strategy had more problems related to Friday and delivered poorer results when Friday trading was enabled – but only slightly. It’s worth noting that in the test I mentioned above, where the end result was a net profit of only 4%, the strategy was up 80% at one point (after the first two weeks!), but fell to -38% in the space of two days’ trading with a string of losers (and neither of those big losing days was a Friday). It then soared again to a profit of over 60% before hitting another big losing day (again, not a Friday) and falling back to finish at just +4%.
The one encouraging aspect of the testing is that I’ve run 25 to 30 tests so far, and all except those for EURUSD during the last month have delivered a net profit. I plan to disable this strategy and focus only on Scalper trades, at least until the EURUSD is firmly in a trend again.
If you have enough PC power to run the tests in a reasonable timeframe, I’d recommend running some backtests of your own with differing combinations of parameters using the optimization feature in Metatrader to give feedback on which combinations work best. And please share your results with the rest of us!
Well, there you have it! Thank you Dave for sharing your findings!
I am soft-launching the forum today [[applause]], and I will sending an email about that shortly. I will post the address to the forum here on the blog today or tomorrow.
Lastly, I will have a little more information on the new VPS I mentioned tomorrow. I can assure you that you won’t run into all of the issues that you are currently facing!
