So he concluded that the Scrypt algorithm somehow gave a coin it's value? That doesn't even make sense and makes his conclusion suspect. Bitcoin dwarfs all other alt-coins combined in terms of pretty much all those factors and does not use the Scrypt algorithm, therefore the Scrypt algorithm has absolutely nothing to do with it.
Here's a link to the episode: https://letstalkbitcoin.com/blog/post/the-bitcoin-game-18-jeremy-gardner-adam-hayes
He looked at "66 commonly traded altcoins against bitcoin". So in his analysis, of those alts, Scrypt algo apparently did better over other algos.
I'm not sure what time period he used, but you know crypto - much can change quickly (see $UNO hash rate). I think he did solid work using the data he had available to him at the time. From the sound of it there was visual info accompanying his presentation - I'd like to see the $UNO data. As it is - I thought the info was very positive for Unobtanium.
I very seriously doubt it had anything to do with the algorithm. Scrypt was the second successful algorithm to come out, so it has the majority of the non-SHA256 traffic and really the only one to have ASICs built for it. I can in no way see how the choice of algorithm makes a single tiny bit of difference to the miner, as miners will mine whatever is the most profitable with the hardware they have. I doubt many of the people (if any besides myself and a few others) in this thread can even describe any of the algorithms or how they work, so why would they care? They don't.
Anyway, that'd kind of off topic. I think he should have left the algorithm bit out of the whole thing, as it is a red herring and data puffery.
Don't be confused - all this points to (scrypt vs SHA) is this: the harder it is to mine an otherwise identical coin, the higher its cost of production. Factors that tend to increase cost of production therefore increase the exchange value (as evidenced by the data and statistical analysis).
In terms of the data the t-statistic was very highly significant (p < 0.0000) for that variable.
It is comparing apples to oranges with ASICs, but the point is that the higher the cost of production the higher relative value.
