3 ms·

Both the article and https://www.titanvest.com/performance/ https://www.titanvest.com/performance/ cite performance relative to S&P 500. If you expand out discl

Both the article and https://www.titanvest.com/performance/ https://www.titanvest.com/performance/ cite performance relative to S&P 500. If you expand out disclosures it says "Figures cited for 2017 and since 2004 represent backtested performance of a hypothetical account using Titan’s investment process, not an actual amount." It's trivial to overfit a backtest to get whatever results you want. For a startup whose objective is to "enable you to become a better investor" they might want to start by explaining why you shouldn't take their performance page seriously.

AznHisoka 8y ago
yep and it makes this statement rather suspicious:“Of the ~3500+ hedge funds out there, we track ~5% of them. We believe these are the good guys: long-term oriented and rigorous in their research.” Its the 5% that performed well in their backtest, but that doesnt mean they will perform the best going forward. What Titan did in 2016 was chose the top 5% of funds that did well up until that point. Then when they saw it performed bad in 2017, they most likely went back and chose a different 5% that did better. And recalculated all the returns. The problem is you cant keep doing this once you actually start investing with real money.
- norswap 8y ago
  Do you have any evidence for any of this, or did you just pick the most uncharitable possible scenario and present it as factual?
  - AznHisoka 8y ago
    This is just all theory, of course but this is something most backtesters do, so I wouldnt rule it out. It helps to be overly skeptical when anyone claims to beat the index. Because it is such a powerful claim.
    - neonate 8y ago
      You said "What Titan did". That's a far cry from "I wouldn't rule it out".
  - sseveran 8y ago
    They should disclose how many backtests were run along with the methodology for evaluation. Backtests should be very rarely run.

AznHisoka 8y ago

yep and it makes this statement rather suspicious:“Of the ~3500+ hedge funds out there, we track ~5% of them. We believe these are the good guys: long-term oriented and rigorous in their research.” Its the 5% that performed well in their backtest, but that doesnt mean they will perform the best going forward. What Titan did in 2016 was chose the top 5% of funds that did well up until that point. Then when they saw it performed bad in 2017, they most likely went back and chose a different 5% that did better. And recalculated all the returns. The problem is you cant keep doing this once you actually start investing with real money.

norswap 8y ago
Do you have any evidence for any of this, or did you just pick the most uncharitable possible scenario and present it as factual?
- AznHisoka 8y ago
  This is just all theory, of course but this is something most backtesters do, so I wouldnt rule it out. It helps to be overly skeptical when anyone claims to beat the index. Because it is such a powerful claim.
  - neonate 8y ago
    You said "What Titan did". That's a far cry from "I wouldn't rule it out".
- sseveran 8y ago
  They should disclose how many backtests were run along with the methodology for evaluation. Backtests should be very rarely run.