Time match discussion

there's been some dissent about time matches recently in the revision thread and i was of the opinion that this topic really deserved its own thread for discussion, so here it is.

FalcoXII wrote:

I've always been skeptical of time matches, as I wonder what prevents fast characters from running away after an early lead.

and this is what got me thinking, although the issue is a lot bigger than just plain camping mentality or superdefensive gameplay. here's the logic as i see it:

1. in time matches, the players are no longer the deciders of when the match ends; there is now a third party (the countdown clock) under whose restrictions they have to operate. in simpler terms, the game ends in 5 minutes whether the players like it or not.

2. because of this limit, the ultimate deciding factor of the game is now based on who can score the most points in the allotted time limit. since this is the END GOAL of the game, the efficiency of scoring points is now the single most important factor in a player's performance: a player who scores one point per minute is better off than a player who scores 0.5 points per minute.

3. player performance in SFA is significantly influenced by character speed, character health, and character specials, all of which serve to either expedite players' aggression or hamper their opponents' aggression (and often both).

4. of these qualities, speed is the most influential in deciding points-per-minute EFFICIENCY.

CONCLUSION: there is no reason to not play a faster character in time games.

to put this in simpler terms, when the game becomes all about how many points you can get and how fast you can get them, the players who can do more stuff in less allotted time have an advantage.

the point up for debate here - and this is certainly up for debate, so please chime in - is whether point 4 is actually correct. is speed actually the most important factor in deciding points-per-minute efficiency?

as the guy mostly responsible for bringing Falco out of obscurity in the Vancouver league, i have to say it probably is. there's a very strong feeling of security, of ease, when you play Falco nowadays; you know that you're never going to be strapped for time, that you can get away with not planning ahead as much, that you can afford to make some mistakes because your speed will allow you to get out of them later. of course, this could be because there are just a lot of top players playing fast characters right now, but doesn't that by itself indicate something about the metagame's state at the moment?

i think the reason for this is that health, on its own, is not an attribute which automatically grants a user any particular freedom on the battlefield. in time games, your health doesn't get you the efficiency which speed gets you. it doesn't get you the ability to recover from mistakes, because even if you survive hits which would kill a more fragile and faster player, you can't be patient and wait to patch it up, regroup, buy time - because you don't have time any more. all health will do is save you in the actual act of combat, once or twice. meanwhile, once a faster player does damage to you that you can't reciprocate, your advantage is gone. specials are more or less the same way - Krys's barriers are, more or less, just an extension of her health; Pep's CB isn't, again, on its own an attribute which grants efficiency. in the latter case, you might get a lot of control over a very specific area and it might save you in mid-range combat but when your Falc opponent is up a point and there's a minute left in game, that CB is not going to grant you any boons that you aren't already given by your own skill as a player - so why would you pick Pep in this scenario?

when you play a slower character, you're playing chess while your opponent is running a race: how many weapons can i pick up in the given time span? how much can i rush without penalty? having control of the map in SFA is a big, big deal - knowing where you're allowed to run, where you're allowing your opponent to run, looking for opportunities to keep someone out or let yourself in - and in time games, there's an additional restriction on that control, because if you can't make the required pushes in time, you lose. to that end, i propose that no player should be barred from making the pushes they need to make just because of their character selection.

all of this comes with a caveat, though, and it's this: i really like time games! we've been playing them through less than half of the league's existence and we've already had uncountable clutch moments, hair-raising 30-second pushes, all sorts of great plays, interesting scoreboards, a dearth of total shutout games, and, notably, we can actually play tournaments and know how long they're going to take now. games are often more exciting in time format and i think that's great for a competitive scene. and time games do have great advantages: apart from limiting camping and boring gameplay, they force players to think faster, to make snap decisions, to look down the road a little further and try and see where they're going to be in a few minutes' time, not just after the next point or next missile pickup. so i'd be very sorry to see time games go completely, and i wouldn't want that to happen - but let's not discuss what should happen to the formats yet, since we haven't even had any discussion on the matter itself.

of course, this is so far theoretical, but i do have some numbers regarding stats in the time era which are interesting. i'll put them up in another post.

now here are some numbers. a good first question is whether fast characters are getting the best of all other matchups, so here are the win rates for Falco, Krystal and Fox ever since the time era began (from set 7 onwards, comprising three tourneys: the Gathering, Floss 1, and Kelowna 2012).

here are the matchups for the Gathering and the first set - a pretty irritatingly small sample size, but there you have it. these are all the matches in which a "fast" character was involved (Floxtal) playing a character slower than them:

FvP: 1-1
FvS:
FvX: 2-1
FvK:
XvP: 1-0
XvS:
KvP:
KvS: 3-0

okay, not a big deal; the KvS distance has always been pretty significant and people have been playing Falc well since the game started. here's the next tourney included - Floss 1 - when people started to play a little faster and Akp (among others) made the switch to full-time Falc:

FvP: 2-1
FvS: 3-2
FvX: 3-1
FvK: 2-1
XvP: 1-0
XvS:
KvP: 0-1
KvS: 5-1

now we see something a little interesting: apart from anything else, that 3-2 spread in FvS would never, ever have happened in the points era. Falc is pretty firmly in the lead here; the other matchups don't tell us too much except that going Slip vs Krys is probably not the best option... but you knew that. finally, here are the numbers from Kelowna 2012 added to everything previously:

FvP: 2-1
FvS: 3-2
FvX: 3-1
FvK: 2-2
XvP: 1-0
XvS: 0-2
KvP: 1-1
KvS: 6-2

no change in the Falc numbers other than another Krys win (Ganon vs STG on Katina i believe). the XvS matches are from Rambler vs STG.

maybe these numbers aren't very illuminating by themselves, so here's some more insight. there were 63 total matches in this period (of all matchup types). there are 15 possible matchups in the game (5c2 + 5, accounting for ditto matches), so Falc dittos account for 6.6% of the possible matchups that could occur.

in these 63 matches, though, 13 of them were Falc dittos. 13! that's a little over 20%. where did that extra 14% come from? are people so scared to pick anything but a character that can keep up with their opponent's mobility?

one final set of numbers: Cac suggested that the previous numbers were "just on the very edge of being acceptable" and that a more specific look at who's playing what character might be necessary. we totaled the number of end-of-set matches entirely - those matches which were the third game in a set or overtime. the idea here was to get a picture of the high-stakes games in their later stages: what do the heavyweight players pick when they really, really want to get the win?

we pooled 30 total character picks from the three tourneys, comprising all the third and second games in sets as well as any overtime or sudden death matches. of these 30 picks, Falc was picked 13 times; when we looked solely at the two most recent tourneys (Floss 1 and Kelowna 2012) the number diminished to 9 Falc picks out of 20 total picks: again just under 50% but a little higher than the previous amount (more precisely, 43.3% and 45% respectively). when you consider that there are 5 playable characters and that Falc comprises one of them, this is a weirdly lopsided number, more than twice what it "should" be under a totally even game...

of course, this could mean anything. i don't think there's a terrible slant and everyone is going to pick all Falc all the time based on anything here, but it is certainly right on the edge of acceptability, as Cac said. we also tested some matches today in which i auto-locked Falc all the time, one in which his Slip took me out on Simp2 handily; i wasn't playing my best but it wasn't as if either of us were particularly getting gimped. i did later get him back on Katina in that same matchup; it's not any indication of rampant exploitability.

a better preventative measure (to, admittedly, a problem which may or may not be there) which Cac tossed out was to disallow players from picking the same character twice in a set. of course in 1-game matches this won't change much but it would certainly stop all-Falc tourneys from creeping into the normal metagame and would give non-diverse players a reason to branch out a bit. i certainly like this idea better than going back to points exclusively or even allowing anyone to choose their gametype.

now give me some opinions

The question is, did the metagame evolve to fast characters out of necessity. In other words, if we switched back to point would the metagame evolve back to using more slow characters.

I don't know. I find time to imbalanced, personally. This could be an intense debate.

I never did like the idea of time matches, I always play point or survival. Occasionally we play time for tie breaks, or two life survival. Unless the slow character has the right items (stealth, weapons and /or jet pack) then they often can't stop the "get a kill and run" strategy.

it sure would help if i could look at some of your guys' games. is there any possibility of uploading some sample matches - from any of you? i'd really like to see what's going on, how everyone else works under the limits of the timer, and what the faster characters are doing that's so exploitable, because in my experience it's pretty damn hard to guarantee a win just by playing defensive runaway games. in other words, i want to know what your opponents are exploiting that we aren't; does anyone think they could get some recordings?

HAXage wrote:

it sure would help if i could look at some of your guys' games. is there any possibility of uploading some sample matches - from any of you? i'd really like to see what's going on, how everyone else works under the limits of the timer, and what the faster characters are doing that's so exploitable, because in my experience it's pretty damn hard to guarantee a win just by playing defensive runaway games. in other words, i want to know what your opponents are exploiting that we aren't; does anyone think they could get some recordings?

Me and those I play with take special items off and i'm pretty sure that's why the faster characters can just run after the first kill. I'll try to get a video of special items on and off matches.

that would make a considerable difference, yeah, although i don't know if missiles are the sole reason nobody tries to abuse speed in that particular fashion. i'd just really like to see for myself what's going on; it's a lot easier to parse actual footage of games than it is to try and make sense of a description.