What is the "typical" win rate?

[KenLBlack]

What is the "typical" win rate?

I have to admit, after playing this game in earnest for about two or three months, purchasing about 150+ kegs and making it up as high as roughly 3400 mmr in ranked play, it's getting pretty tiresome losing, having dived the past few weeks to under 3000 again! What is a "typical" win rate in ranked play in the circa 3000 mmr level? Mine right now--after losing 16 of my last 20 matches--is around 36% this season only. Is that normal or am I just a completely awful player with hideous run of bad luck?? It's not that I don't have decent cards--I think I do--but it's hard to see what a "typical" win rate is using the tools available. The one in Gwent itself only allows you page through the top-ranked players, and I don't believe it's possible in Gwentup, either.

I've often been one to try to see the bright side and to keep plugging along knowing that it would eventually get better--and it has to a great degree. Heck, I made it almost up to 3400 until I started diving for the past two weeks, and of now I back under 3000 at around 2900. So I ask....what exactly is "normal" for a win rate?

Thanks!

 

[TrompeLaMort]

By definition, a typical win-rate is 50%. But actually, since in Gwent you can rank up with <50% win-rate until ~3.9k, I would say the typical win-rate at 3-3.4k is probably ~45%.

In any case, half the people don't win more than 50% of the time, it is life. You shouldn't let this get in the way of you having fun, and try to improve your gameplay.

 

[Jeydra]

36% win rate at 3000 MMR is very unlikely to be due to variance. Top players in this bracket are likely to win 75%+ of their games, possibly more.

This doesn't mean you are completely awful however. It could also be that you're not playing a good deck.

 

[KenLBlack]

Thanks for your replies. Being new-ish, I rely some on netdecks but think (well...thought) I had the overall strategy of things down. I certainly realize trying to copy a deck or two is not enough and that one has to know how to play it skillfully. I've been playing a deck similar to this one (https://www.gwentdb.com/decks/44358-1st-place-wild-hunt-lan-axemen ) and also this one (https://www.gwentdb.com/decks/45441-aenimas-eithne-control-4100-mmr ) but have been having incredibly bad luck with both as well as Henselt Machines. I'm trying to not get too tied into developing just one deck, since it's good to learn about the factions by, of course, playing them. I've re-watched the last two Gwent Opens online, as well, to pick up some tips from the casters as they go over the matches.

Ugh. Well...keep plugging away, I guess. Thanks again for your replies.

 

[liezldiezldee]

I have to admit, after playing this game in earnest for about two or three months, purchasing about 150+ kegs and making it up as high as roughly 3400 mmr in ranked play, it's getting pretty tiresome losing, having dived the past few weeks to under 3000 again! What is a "typical" win rate in ranked play in the circa 3000 mmr level? Mine right now--after losing 16 of my last 20 matches--is around 36% this season only. Is that normal or am I just a completely awful player with hideous run of bad luck?? It's not that I don't have decent cards--I think I do--but it's hard to see what a "typical" win rate is using the tools available. The one in Gwent itself only allows you page through the top-ranked players, and I don't believe it's possible in Gwentup, either.

I've often been one to try to see the bright side and to keep plugging along knowing that it would eventually get better--and it has to a great degree. Heck, I made it almost up to 3400 until I started diving for the past two weeks, and of now I back under 3000 at around 2900. So I ask....what exactly is "normal" for a win rate?

Thanks!

80% to 4000 and then ~60% for me...

are you playing a top tier deck or something you made yourself?

 

[Void_Singer]

The way the MMR system works in gwent, there is an MMR spread, and an rank modifier. the spread determines the base amount of points to be won or lost, and the modifier is used to generate extra points at lower ranks (by multiplying the winning value, and dividing the lowing value). Assuming the modifier is 1 at rank 20, it would take a >50% win rate (of games played at R20) to make it to R21.

The (grossly simplified) formula would look something like

Rank_Modifier = 20 / Your_MMR_Rank; //-- not your highest rank but the rank you'd be at for your MMR
Base_Points = ABS( Player1 - Player2 ); //-- actually this would be K-factor but we're trying to make this easy
if (Win){
  Your_MMR = Your MMR + (Base_Points * Rank_Modifier)
}else{
  Your_MMR = Your MMR - (Base_Points / Rank_Modifier)
}

 

[KenLBlack]

80% to 4000 and then ~60% for me...

are you playing a top tier deck or something you made yourself?

Well, I thought they were top tier, liezldiezldee , but perhaps not anymore. You have a very impressive winrate, by comparison! (See my two specific links in my earlier reply for an idea of some of the decks used.) Perhaps it IS my brain power; or lack thereof! In full disclosure, I am older than your "typical" gamer at age 58, so perhaps a bit of mental nimbleness is lacking, but I've tried to hold my own.
​​​​​​​
Void_Singer , you never cease to amaze me with your observations and your insight into how things work. Your work on the card reference is outstanding, and that is an interesting insight into the way the MMR works. I did know that it gets much tougher as you move up, though I guess I've been unpleasantly surprised.

 

[devivre]

Axeman can be very strong, but it seems to require experience to be good at it. The other deck you posted also looks like you can mess up a lot quite easily.
Something like Greatswords Crach or a typical Brouver Elves deck is definitely easier to play.

 

[ZenaRose]

My winrate is 955-734-30 its 57% for ~1700 games. Im no tryharding, playing only Shupe decks and often concede even in ranked if i dont want to play vs toxic deck like brouver-cleaver combo

 

[Jeydra]

Thanks for your replies. Being new-ish, I rely some on netdecks but think (well...thought) I had the overall strategy of things down. I certainly realize trying to copy a deck or two is not enough and that one has to know how to play it skillfully. I've been playing a deck similar to this one (https://www.gwentdb.com/decks/44358-1st-place-wild-hunt-lan-axemen ) and also this one (https://www.gwentdb.com/decks/45441-aenimas-eithne-control-4100-mmr ) but have been having incredibly bad luck with both as well as Henselt Machines. I'm trying to not get too tied into developing just one deck, since it's good to learn about the factions by, of course, playing them. I've re-watched the last two Gwent Opens online, as well, to pick up some tips from the casters as they go over the matches.

Ugh. Well...keep plugging away, I guess. Thanks again for your replies.

If you're playing these decks card-for-card then, sorry to say, your playskill is the deciding factor because they're strong enough to get you to 4k++ MMR. For example Axemen, according to the last meta report (https://gwentup.com/report/18/12), is comfortably above 50% win rate. The list you linked is also a strong one. The ST deck you linked is not so strong - I can't approve of handbuff cards without payoffs, in particular where are the Swordmasters? Plus, Farseer + Scorch is not a good combination. But it's focused enough that you should be able to reach 4k MMR.

If you like, you could describe a game step-by-step and I should be able to help you with your plays, especially with Axemen since I'm very familiar with that deck.

 

[KenLBlack]

Thank you, Jeydra - I appreciate the offer! Part of my issue is luck of the draw--not pulling many gold cards. But IF I'm lucky enough to pull a gold weather, I will often start with that, because the Axeman is otherwise a sitting duck for anything with an 8+ ability to kill without being buffed almost immediately. I have thus far been killed by Iorveth or a buffed Braen when playing against SC, anyone with Alzur's Thunder, and consumed by a Kayran playing against Monsters. (And don't get me started on the Viper Witchers.) That is also the issue with Derran, who has also been Mandraked more often than I care to remember.

I assume the main key to winning is buffing up the Axeman or Derran, either via weather damage or harpooning the opposing player into a Skellige storm.

Basically, I would say no matter what I lay out, it gets taken off of the board virtually immediately, so I therefore have to rely on the whaler or a ship doing damage, but the effect is nullified considerably if there is not an Axeman sitting out on your field of play.

Thanks for any insights ahead of time. And thanks for the tips on the SK deck. I will have to do something about that, as well.

 

[jgolden234]

Once I get to rank 20 I find I go on win streaks that are quickly out done by losing streaks. It is very frustrating. They are definitely making me earn 21!

 

[KenLBlack]

Thank you SO much, Jeydra for taking the time to write such an extensive guide! I greatly, greatly appreciate it! I will try to digest all of this and see how it goes. Fortunately, I've been doing some of these already, though still without success. (And I'm fairly sure I have the concept of mulliganing/blacklisting down since I've read a few articles online on it, but I will double-check.)

Thanks again!!!

 

[KenLBlack]

Jeydra, if I could give you another "REDpoint" I would. Thanks to your assistance, I just got four wins in a row...one in casual (just to test!) and the next three in ranked. I don't think I've ever pulled off four wins in a row before on any level. Thanks again! I had read that the Gwent community in general is quite friendly to one another compared to some other online games, and I am delighted to see that this is true! I guess partly given my age, it helps to see things written down combined with a YouTube video or two as opposed to just watching the videos alone. GwentDB can be inconsistent with the deck guides. I would highly encourage you to write some guides somewhere if you haven't already; you have a knack for it!

 

[Draconifors]

Moved the latest posts to a new, separate thread Tips for playing Axemen, as they really were getting a bit far from the actual topic of this thread.

 

[Khreygond]

Just want to add that skill is also something gradually earned and when I say gradually I mean VERY gradually. I've got 1700 hours into the game and I would only describe myself as good. I've never reached Rank 21, even though I sit at 4.1k MMR every season.

Getting used to the deck you play is also crucial. copy pasting the best deck in Gwent will serve you right for reaching rank 20 but after that point, people make very few mistakes and anticipate what you are packing. That's when skill becomes the crucial factor. Up until that point, you could be using Tier 2 decks.

I second the thoughts about Axemen. They are, if played right, a very dominant deck. It's what I was using the past few days But if they get too popular, the meta will adapt to shut them down.

 

[hellknight55]

This is obvious, but in order to win, your opponent has to lose. There is a 0 sum gain in this respect. However, as some have mentioned, you can still gain MMR with less than 50% win rate due to the specifics of the system.

So, in games where there is little to no luck involved, the more skilled player should win the vast majority of the time (95%+). Example: Starcraft
Gwent has luck embedded in it though, in the form of randomness (what you draw, what your opponent draws, the specific match-up). So, you while you would expect the more skilled player to win most often, you cannot guarantee it due to luck and match-ups.

So, if luck was predominant, you would expect a standard bell curve / normal distribution for wins. I don't think this is quite the case with Gwent, so let's just say that we shift that bell curve a bit. So instead of the mean being around 50%, let's say a skilled player has a mean of 60%, and a less skilled player is between 40%. So you would expect them to fall within two standard deviations of this mean. That means a skilled player can expect to win between 30 - 90% (2 std. deviations is +/- 30 points). In other words, a skilled player can expect to win 3/10 games at worst, and 9/10 at best.

Finally, a less skilled player is also expected to fall within 2 standard deviations of their mean win rate, meaning 10 - 70% range. So at worse, they will only win 1 out of 10 games, and at best they will win 7 out of every 10 games. You can debate where the mean should be for any group of people, or even person. However, I think it's fine to use a standard distribution model due to the nature of Gwent, in that it has a decent amount of randomness between what both players draw, and of course matchups and maybe even the coinflip lol.