Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Morghas]

So...

My dear friend and I are setting up a spread sheet to record all the games we play so we can try and gather stats for informative, jovial and bragging purposes. This does add a good 5 mins to each game we play as we compile the info but it is actually quite a fun process and a good post-game strategy ponder opportunity. Anyway, I thought I would pose the idea here to see if any of you have any better ideas as to what stats we could gather and more importantly how we could interpret them. Maybe someone has already done this too?

The info we are recording is:
Score
Whether there were Shelters or Colonies/Platnum
Who went first
Amount of turns each
Cards in kingdom (including Events and Landmarks)
5/2 splits
First 2 buys
Empty piles at end of game
What strategy we think we went
The card we thought was cutest...

I dont think there are more hard stats to gather...

Where it gets more interpretive, we are also rating each card we used in the kingdom based on the strategy we had in favour. So for example in my deck the most crucial cards IMO were Trade Route, Treasure Trove, Vault, Nobles, Great Hall so I put Trade Route at 1, Treasure Trove at 2, Vault 3 etc...  My mates were Native Village, Trade Route, Page, Stone Mason he does the same in a seperate row. It doesnt matter that he only has 4 to my 5 cards and despite my strategy being considerably worse I still rate the cards I thought were more crucial to my strat during the game. The cards will then be given a share of x points, distributed something like this:
1: 20 pts
2: 15 pts
3: 10 pts
4: 5 pts
5: 3 pts
6: 1 pts
7: 0 pts
Below 7 is most likely too bad to be scored.
We are not totally sold on point distribution so would love any feedback in that department? Would also be keen to hear whether you think this system of card rating has any value? Obviously after we play a few 100+ games things will begin to get more interesting.

We have another page which lists the cards and shows how many times they have each been seen and what we scored them.

Stats I am then keen to see will be:
Amount of games won by each player when x card is in on the table.
Top 10 kingdom cards based on our scoring system.
How often 5/2 split wins.
How often going first wins.
Who wins more often when shelters, colonies, potions, attacks are available.

Anyway, that is the vibe...

What do you think?

[Qvist]

Wow that's a lot of effort, but I like it because I like stats.

A different approach that came to my mind for the distribution of points is so that you make sure an equal amount of points get distributed each game. So for example if Cultist/BM gets played Cultist gets all available 20 points for example. But if an engine gets played where a lot of cards were important the most important card gets like 8 points, the second most important 5 points, and so on - dependant on the importance - until the total is 20 points. With a fixed distribution you really can't tell how dominant a card was. The second most dominant card gets always 15 points, no matter if it was nearly as important as the most important one or if it had barely any importance. Also evaluating the average (for example 50 points out of 5 games = 10.0) is more significant when you know you have distributed 20 points each game. I also would only score for the winning player because if I play Scout BM, Scout would get 20 points, but it is saying nothing about how terrible that strategy was. Just my 2 cents.

[Morghas]

Thanks for the reply Qvist!

We originally thought that we didn't want to spread the points out like that because in your example cultist would then fly ahead by so much, whereas if you had fixed points at least the top card of every game gets an even rating. You have made me rethink that though because maybe the fact that cultist could win without any other cards shows how strong it is? I think you are right in just getting the winners pick for rating the cards.

We have, however, come up with another idea... Inspired by this article http://forum.dominionstrategy.com/index.php?topic=13822.0
Maybe we could give all kingdom cards a grade instead? From A to whatever maybe D, then if there were 3 equally important cards they could all get A's instead of having to pick out just one card for top place or probably more notably a bunch of helpful cards could all get B's. It also means the bottom cards get a bit of statage even if it is just confirming their bottomness. Maybe something like:

A: Crucial to winning strategy! The strat was built around this/these cards.
B: Crucial support card to winning strat. Played an important role but would be replaceable in another kingdom.
C: Mildly helpful and wouldn't hurt your deck but not needed.
D: Bad! Would negatively impact your deck.

Thanks heaps for the feedback. Yes it will take some time but spread sheet stat gathering is time well spent! Haha! I am most excited about having card champion stats like, whenever Beggar is in play player x wins 95% haha.

Would be keen for any other opinions if they lurk anywhere?

[Polk5440]

Here's my suggestion. (Full disclosure: I don't record anything from real life games and prefer 3+ players and more casual games when I do play in person.)

Definitely record:
Who won
Whether there were Shelters or Colonies/Platinum
Who went first

If you want more detail, record:
Score
Cards in kingdom (including Events and Landmarks)
Empty piles at end of game & Victory condition
** Final Deck Composition
** Clarifying Note

Less keen on:
Number of Turns
5/2 splits
First 2 buys
What strategy we think we went
The card we thought was cutest
Crucial card points

When I review online logs, I usually skip all the way to the end and check two things: final score and final deck composition. This gives me a good heads up on what people are doing and a good representation of someone's strategy and differences in very similar strategies. I do sometimes wish there was a list of what is in the trash, too.

So, if you want to keep more detail about a game, that's what I would do. Record final deck composition, and in cases where that doesn't tell enough of a story, add a note (remodeled all Border Villages away in endgame, ran piles down too low, getting early Witch won me curse split, etc.). It seems like this would not take more time than describing strategy and rating cards, is less thinking, and more objective and consistent.

[Morghas]

Hey Polk!

Yeah we have been trialling the spread and ranking the cards is real teadious, sometimes kind of fun but others not.... Deck composition would be easier and could be more telling anyway.

Thanks for the feedback.

[McGarnacle]

What exactly do you mean by which card was cutest?