Quant funds don't trade with computers - in fact, computers trade instead of humans.
> the number of trades you make doesn't determine alpha
No, but it does determine how statistically significant your alpha is. If I make 1 good trade, it might be alpha or it might be a lucky guess. If I make 10000 good trades, it's unlikely to be just a lucky guess.
> certainly no sense to 'alpha becoming beta.'
Trend following used to be "alpha", but now it's considered "beta" - in the sense that everybody can replicate it, and there is no specific "alpha"-based fee warranted for a fund executing trend-following strategies.
> there is no a priori reason that their strategies must or will become public
the more people know about it, the easier the math behind it, and the more broadly it applies, the more chance there is that it becomes "public" knowledge (public in the sense that many industry practitioners working for different funds know about it)
> but compare that to seriously unscalable fields such as venture capital
Well, given that quite a few venture funds are bigger than $10bn [1], I wouldn't call that "seriously unscalable". But in any case, my comparision was to trend following (Winton has about $30bn), global macro (Bridgewater's Pure Alpha has about $50bn), and passive index investing (SPDR S&P 500 ETF is > $100bn).
Quant funds don't trade with computers - in fact, computers trade instead of humans.
> the number of trades you make doesn't determine alpha
No, but it does determine how statistically significant your alpha is. If I make 1 good trade, it might be alpha or it might be a lucky guess. If I make 10000 good trades, it's unlikely to be just a lucky guess.
> certainly no sense to 'alpha becoming beta.'
Trend following used to be "alpha", but now it's considered "beta" - in the sense that everybody can replicate it, and there is no specific "alpha"-based fee warranted for a fund executing trend-following strategies.
> there is no a priori reason that their strategies must or will become public
the more people know about it, the easier the math behind it, and the more broadly it applies, the more chance there is that it becomes "public" knowledge (public in the sense that many industry practitioners working for different funds know about it)
> but compare that to seriously unscalable fields such as venture capital
Well, given that quite a few venture funds are bigger than $10bn [1], I wouldn't call that "seriously unscalable". But in any case, my comparision was to trend following (Winton has about $30bn), global macro (Bridgewater's Pure Alpha has about $50bn), and passive index investing (SPDR S&P 500 ETF is > $100bn).
[1] http://www.forbes.com/sites/alexkonrad/2015/03/25/midas-top-...