What is a "good" rating?

Previous topic - Next topic

tdellaringa

I'm curious what people see as a "good" rating. Generally, looking at the leaderboard, the top players are in the 60's. It seems to me that any players in the upper 40's+ and into the 50's is a good player, and possibly very, very good.

Personally, I feel like if I am in the 30s, I am not doing well. When I cross over into the 40s, I feel like I am doing okay. I'm currently at 41.

Thoughts?

jsh

From my perspective (as a mid 60s level player), I don't really feel challenged by most players below 50 or so, and when I lose to players at those levels 80% of the time it's because of bad draws or turn advantage, but there's a lot of variation on both sides and obviously strings of bad or good luck can make someone's rating be off. Right now there doesn't seem to be a huge difference between a player in the 40s or 20s, at least from my perspective. So I guess a quick tier list for me would be like:

Dominion Experts: 60+, maybe more like 58+
Really strong players: 55-60
Above average to great players: 50-54
Too much variance to tell: 49 and below.

tdellaringa

Ouch, my ego! I could see from your perspective, a 20-40 player is the same. At my level, there's a pretty big difference between the two. Most players I see in the 20s have little idea about strategy, they buy a lot of random actions, they don't try to thin their deck (or do it soon enough). They buy provinces too soon, etc. Where players in the 40s seem to know the game pretty well, and can play outside the core set as well.

It's interesting to get different perspectives.

jsh

Quote from: tdellaringa on 19 May 2017, 03:59:04 AM
Ouch, my ego! I could see from your perspective, a 20-40 player is the same. At my level, there's a pretty big difference between the two. Most players I see in the 20s have little idea about strategy, they buy a lot of random actions, they don't try to thin their deck (or do it soon enough). They buy provinces too soon, etc. Where players in the 40s seem to know the game pretty well, and can play outside the core set as well.

It's interesting to get different perspectives.

I've seen what you are observing, I just haven't noticed players in the 20 range being that different from those in the 40. Sometimes I'll see a player at 47 totally ignore a key card, trashing, etc. On the other hand, I have played pretty close games against people in the 20 range. I suspect this is partially because the client starts new players at 20 and everyone is working their way up.

limetime

#4
I concur with what JSH is saying.
65+: Top 3-4 Really good. Can easily outplay anybody else.
60-65: Top 50 They usually make fairly reasonable (In my opinion as a 65 lvl player) choices. They sometimes make a few mistakes throughout the game but usually no major mistakes. You have to play quite well to beat them.
55-60: These players are fairly decent. They usually choose the correct strategy and play it reasonably well. They tend to either be more inefficient in their building or are worse at endgame than the levels above them. In complex engines level 65 players should be able to beat them about 80% of the time. In simpler games it is closer to 60%
50-55: This is where the skill level starts to drop off. These players often play the wrong strategy or play it quite poorly.
40-50: These players are very iffy. They have to get lucky to win against higher level players.
20-40: Lots of variance.

Sorry if this comes across as arrogant, that is not what i am trying to sound like.

tdellaringa

This is pretty eye opening. I know that I make a handful of either mistakes, or moments of indecision where I have to make an educated guess, because I am not sure what to do. Sometimes this is because I am playing with a card or two that is new to me. (I only recently began playing with expansion cards.) So I get what you guys are saying.

It also seems to me that after the first opening strategies, the mid-game is the place where you can really screw up. I often will make a mistake there I regret - or just make a play I didn't realize would harm me. I expect 50+ players do not do that.

jsh

Quote from: tdellaringa on 19 May 2017, 04:19:54 AM
This is pretty eye opening. I know that I make a handful of either mistakes, or moments of indecision where I have to make an educated guess, because I am not sure what to do. Sometimes this is because I am playing with a card or two that is new to me. (I only recently began playing with expansion cards.) So I get what you guys are saying.

It also seems to me that after the first opening strategies, the mid-game is the place where you can really screw up. I often will make a mistake there I regret - or just make a play I didn't realize would harm me. I expect 50+ players do not do that.

The most common mistakes I see are at endgame! Can't count the number of times I have seen a player throw away a shot at victory by leaving a pile low or buying bad Provinces.

aku chi

Quote from: tdellaringa on 19 May 2017, 04:19:54 AM
It also seems to me that after the first opening strategies, the mid-game is the place where you can really screw up. I often will make a mistake there I regret - or just make a play I didn't realize would harm me. I expect 50+ players do not do that.

I see sub-50 players make a lot more mistakes in the midgame than 50+ players.  Though, from my perspective (60+ player), the midgame of Dominion is generally the easiest to master.  To put it another way, I'm most surprised when my good-match opponent makes an obvious mistake in the mid game, like buying Gold over Laboratory in kingdoms where that's silly (most kingdoms).  There are kingdoms where the midgame is challenging, of course - mostly those with mid-turn gaining.

Early game decisions are ostensibly easier, but they have a huge impact on the rest of the game, and many kingdoms present tough choices (usually between cycling, economy, and attacks).  I think it's tougher to intuit the right choices in the early game; you just have to play a lot to get a sense for best opening practices.  Of course, you can go pretty far copying the opening of a better player!

Late game decisions are the most challenging, IMO, especially in engine mirrors with plentiful gaining.  There are often super-complicated decisions about which piles to lower when; and if, when, and how to green.  This is where I find myself most advantaged against players in the 50s and where I feel most vulnerable against the strongest players.

Martin plays Piano

I think, everything what's written here is correct – so the interesting analysis about what's making the difference between a level 60-player and a level 55-player is really hitting the point.

As a level 50-player I am quite aware how difficult (one can say nearly impossible) it is to beat a player >55 due to their described capabilities (better opening, better mid-game, better end-game) – there is something within those higher levels which could be pictured as a mixture of broad experience (some 1.000 games), mathematical genius and some kind of pure magic.

You can have a look in several Youtube videos with the experts playing –and you might understand, what strategic possibilities the kingdoms are offering, you never ever would have detected. I am convinced that above 55 the region starts where many players never can step in – because this is the region of real talents paired with the ability to play everyday for several hours with raising passion and having a network of other experts to discuss about their strategies.

Important to say, that the component of luck, which is often mentioned as an essential element of dominion games, is more and more powered off in these higher levels – so top players even win against a (weaker), but lucky drawing opponent – their skills are trained enough to mostly turn the tide.

On the other hand the 55+ players are about 250 persons – which shows a 2% portion of the whole player range on this platform. With the 50+ players there are additional 650 persons – so it might be good to know that everyone playing in the big pool of the remaining 90% is having a chance to get into the Top1000, but only with ambition, some talent and some time.

Nevertheless you can say being at level 41 is much better than the overall average here, so I am a good player ... - so finally everyone should follow his own targets to get happy, that's the main thing.

Have fun
Rachmaninoff

Stef

I don't think there is a very meaningful answer to this question... it depends so much on your perspective.

For what it's worth, I think there is a huge difference between level 20 and level 40. Not from the perspective of a level 60 player, who will lose to neither one of them and possibly shake his head when observing the play. But nonetheless a big difference in understanding the game, which will show when they play against each other.

tdellaringa

Quote from: Martin plays Piano on 19 May 2017, 09:35:32 AM
You can have a look in several Youtube videos with the experts playing...

Do you have some links, or some channels to point me to that I could watch? It's a really good way to learn.

Quote from: Martin plays Piano on 19 May 2017, 09:35:32 AMOn the other hand the 55+ players are about 250 persons – which shows a 2% portion of the whole player range on this platform...the remaining 90%...

I was thinking of this last night. I think it's pretty tough to say 90% -  98% of the player base are poor players. Yes the top 2-10% are the masters, sure. Especially with a game so complex.

Having said all of this, Dominion is a really fun game and I enjoy playing it. I'm really happy the online version made a comeback and is improving. Trying to learn to be better is quite challenging and mentally stimulating.

Martin plays Piano

I have no special channel or link for the YouTube stuff ... -

but you will get tons of matches by searching within YouTube for
"Dominion League match"

and you can filter it down to special players like
"Dominion League Stef"
"Dominion League Qvist" or just
"Dominionstreams" and many others

Most games are streamed from the league matches taking place since several years now and there is also a community here on Shuffle IT playing live matches with live streaming and live comments.

Perhaps I am not the correct person to give any recommendations on that, but here in forum there a some players who can give good advice how and when it is worthy to participate as a spectator.

Have fun
Rachmaninoff

tdellaringa

I had NO idea there were leagues. Wow. I'll give that a try.


Ingix

I'd like to remark that the level, as displayed in the game client, is kind of a strange beast. For example, when looking at http://dominion.lauxnet.com/leaderboard/?full=true at this moment, my level is 46.06 and on the same page it shows players with levels from 46.00 to 46.10. I guess most people would intuitively say that all those players have the same strength and the difference is negligible.

But if you look at the μ-value of those players (which represents the (unscaled) strength of the players as estimated by the ratings algorithm, see http://forum.shuffleit.nl/index.php?topic=1679.msg5891 for details),  my μ-value is 0.06, but there are 3 μ-values above 1 and the lowest is -0.17 (all from players with levels between 46.00 and 46.10).

If we use the scaling factor of 7.5 as detailed in the above mentioned post, the difference between -0.17 and 1.31 (the highest value) in μ-value is actually 11.1 levels as expressed in the game client. Which is roughly 100 times more than the displayed range of 46.00 - 46.10 would suggest.

The reason is that the level is calculated not just from μ, but also from φ, an estimated accuracy of the estimate μ. Basically (and for details I again refer to the above mentioned post by Stef) the lesser number of games you have played, the higher the perceived inaccuracy φ. Conversely, when you have played many (hundreds) of games, your φ becomes small.

Since the level is calculated as 50 + 7.5 * (μ-2φ), it will show a player with μ=1.20 φ=0.86 (13 games played) as having the same level (46.06) as myself, with μ=0.06   φ=0.29   (79 played games).

The problem I see is that that person with μ=1.20 is *clearly* better than me. That the rating algorithm is more unsure of the exact strength of that person is clear, 13 games is not much to go by, but putting it on my level is (at least to me) obviously wrong.

The stated reason for using a scaled version of (μ-2φ) is (quoting from the above post)

Quote
Glicko2 claims it's 95% certain your actual skill is between (μ-2φ) and (μ+2φ), and suggests using (μ-2φ) for rating / leaderboard, so we're doing that.

I understand that if you have some entity that objectively exists and you want to estimate if some person meets a certain quality level for that entity you want to err on the side of caution. In that case using a 2.5% of error quantile is reasonable.

But while there is obviously a skill cascade involved with Dominion, I think that the concept of a 'pre-existing' numerical skill level that the Glicko2 rating is estimating is not valid. So I think that using (μ-2φ) is a bad band-aid that declares players of widely differing skills as "equal levels".