Making games more unpredictable

[Spunkmeyer]

QUOTE (Cort @ Dec 30 2010, 11:49 AM) That. QFT.

If Baker's data is correct (which I don't question), my interpretation is that it proves both that Allegiance is highly player-skill based (yay!) and that AS correctly reflects refplayer skill (yay!).

The second part of that does not necessarily follow unfortunately. I assure you if I created a new account from scratch, I could achieve a very, very high AS rank. Most of it has to do with game selection, nothing more.

[pkk]

QUOTE (Spunkmeyer @ Dec 31 2010, 12:53 AM) The second part of that does not necessarily follow unfortunately. I assure you if I created a new account from scratch, I could achieve a very, very high AS rank. Most of it has to do with game selection, nothing more.

AllegSkill tries to rank a player from first game on someone plays. And that's the problem.

New players don't get a high rank, because the first games effect the final rank more than games played after the first 50-100 games. Of course they suck, because they learn the game...

The older games should effect your rank lees than they currently do.

That means I will loose my high rank (Yes, I suck!) and others, which sick at low ranks, because they had to learn the game the hard way (from loosing games) will get a better rank, which would make ranks more predictable.

[Cortex]

QUOTE (Spunkmeyer @ Dec 31 2010, 12:53 AM) The second part of that does not necessarily follow unfortunately. I assure you if I created a new account from scratch, I could achieve a very, very high AS rank. Most of it has to do with game selection, nothing more.

*sigh* You don't pay attention for once and miss an "on average" and suddenly you're in the middle of Yet Another AS Discussion (which is neither what I wanted, nor do I care the least bit to have that once again ). Yeah yeah, the system can be cheated, not really my point.

Maybe we should forge an Alleg rule out of this: any sufficiently deep gameplay discussion is indistinguishable from an AS discussion.

Or make it Godwin's Alleg law: "As an Alleg discussion grows longer, the probability of a it being turned into an AS discussion approaches 1." (But I'm sure that has been said before as well.)

[Spunkmeyer]

QUOTE (pkk @ Dec 30 2010, 06:21 PM) AllegSkill tries to rank a player from first game on someone plays. And that's the problem.

That, and on top of that it can't take other variables (command, settings, balance, rock distribution) involved into consideration, so it at best hopes for high correlation.

I do agree with the overall sentiment expressed here though, Allegiance must be a game of skill, not one of luck. We are not quite there though and we'll never be there as long as the current "rock lottery" stays in place.

[Broodwich]

none of this thread even belonged in cc forum anyway, should have been moved to gameplay

[Spunkmeyer]

QUOTE (Cort @ Dec 30 2010, 06:46 PM) Or make it Godwin's Alleg law: "As an Alleg discussion grows longer, the probability of a it being turned into an AS discussion approaches 1." (But I'm sure that has been said before as well.)

Sorry Cort, AS is relatively new for me having been away so long, so I guess I like talking about it

[Drizzo]

Rock distribution has less of an impact on game outcome than some people make it out to be. Maybe in squad games yes, but in PuGs, Sup is 100% viable, and in most cases, better.

[Spunkmeyer]

QUOTE (Drizzo @ Dec 30 2010, 06:54 PM) Rock distribution has less of an impact on game outcome than some people make it out to be. Maybe in squad games yes, but in PuGs, Sup is 100% viable, and in most cases, better.

I agree. But like you said squad games are very exp focused, and there is tac in the mix too. But the placement of the rocks is important too - I have won/lost a lot of games based on how easy or difficult the tech base was to cap/bomb.

[Drizzo]

The one thing I loved about phoenix was having a giant flying bios sup.

[Dorjan]

Rock Band is fun