JKTKops#64353
My feedback
2 results found
-
47 votes
An error occurred while saving the comment JKTKops#64353
supported this idea
·
-
45 votes
An error occurred while saving the comment JKTKops#64353
commented
I've just played a 45 minute match in which my opponent played a control deck with no win conditions and played it so slowly that they eventually lost to the 30-minute match time limit. This behavior in any paper event would result in a warning (IPG 3.3) and if the behavior continued and the judge believed it was intentional, a disqualification (IPG 4.7). Given the decklist, which I checked with The Stone Brain, a disqualification in paper would be inevitable. This is the **first** match I have ever seen any player's match timer go below 12 minutes. To me, this is a glaring red sign that the match timer and time control system do not respect how (honest, reasonably-paced, and fair) players actually play the game.
Remember that on Arena, we don't have to shuffle our decks to draw opening hand(s) and the vast majority of situations with many triggers/replacement effects/confusing continuous effects are handled automatically and instantaneously. So why are the match timers ~10 minutes per player longer than most in-person events?
As I commented below, the time control system for turns is ineffective and should be replaced. But with this comment I aim to point out that the time control system for entire matches is also inappropriately tuned and is much easier to fix.
JKTKops#64353
supported this idea
·
An error occurred while saving the comment JKTKops#64353
commented
The developers of the now-abandoned [Prismata](http://prismata.net/) apparently put a lot of effort into understanding and designing time control systems that reward both good play and good sportsmanship. They [blogged about time control taxonomy](http://blog.prismata.net/2014/06/17/the-solution-to-frustrating-time-controls/) after implementing it, though their taxonomy is strangely lacking byoyomi time controls, which is close to what MTGA uses today. The primary difference is that MTGA today does not seem to have a clear duration for the main time, and has an increment to the main time every time you take a game action.
Prismata's time control consists of two clocks, your "time bank" and your turn time (called the "increment" in the blog post). When you pass the turn, 20% of your unused turn time is added to your time bank and then your turn time is reset. On difficult turns, you can run out the turn time and go into your time bank. As a result, no unused time will be added to your time bank at the end of that difficult turn. If your time bank ever runs out, you lose immediately.
Having played Prismata for several years before it was abandoned, I can say empirically that that time control works incredibly well. The longest you ever had to wait to claim a win after a ragequit, on the slowest time control, was about two minutes. (On Arena, if your opponent has zero timeouts, you appear to have to wait approximately 4 minutes currently.) Prismata is a faster-paced game than magic and so the 20% number was appropriate. Some calibration would be helpful for the MTGA environment, to support both combo decks that need more time for one long turn and to support the mobile environment where legitimate crashes and reconnection time are a concern.
I recorded the amount of time I spent on my mulligan in my last 5 matches. The average was a bit over 16 seconds. The mulligan is the most difficult decision that usually has to be made before turn 3, so it's a good indicator that the turn timer should probably only be about 30 seconds long. Perhaps it could grow to 60 seconds on turn 4 or 5, or grow slowly over the course of the game. Either way, combine that with a 60 or 90 second initial time bank and use, say, 50% saving rate instead of 20%, and your typical combo loop of 3-5 game actions per point of damage will have access to approximately 2.25 minutes of time on turn 4. In my experience, this is plenty, although I've heard from others that they want to be able to click through their loop at less than the fastest possible rate (I have no idea why).
Acererak combos specifically are a different breed of problem and I don't think MTGA should be trying to solve that problem with the timer. That combo requires 10 game actions (15 if anything is giving Arena a reason to ask you to manually resolve triggers) per single point of damage, or 200 (resp. 300) game actions to win. Existing Pioneer Lotus combo or Historic Lotus Breach decks are the next worst offenders IMO and in their worst case, need a bit over half of that to win. The workaround to the Acererak problem for now is probably to play a Greenbelt Rampager combo instead; it's the same combo, slightly less reliable that you'll have the colors of mana you need, but takes a quarter of the game actions to win. A longer-term "fix" for Acererak and similar combos is a loop recorder. I've discussed extensively why arbitrary combo detection is not feasible for MTGA but a loop recorder is fully within reach. To prove my point, someone has even implemented a simple external one and shared it on Reddit.
But first, I'd like to see motion to fix the timer, and a timebanking system is a much better match to how players actually play the game than a byoyomi system. This would disincentivize roping, and in the situations where players rope anyway, drastically reduce the amount of time required before a win can be claimed.
.png){kind=link}
If this happens in BO3, you can't sideboard (even if you alt-f4 and rejoin the match) and will lose the match due to timeout.
How is this still in the game? It's been 6 years.