Recommend
 
 Thumb up
 Hide
29 Posts
1 , 2  Next »   | 

BoardGameGeek» Forums » Everything Else » OLD BGGBlogs (do not use)

Subject: Which List Best Represents "Best Games of 2004"? rss

Your Tags: Add tags
Popular Tags: [View All]
Randy Cox
United States
Clemson
South Carolina
flag msg tools
designer
1024x768 works just fine - Don't Wide the Site!
badge
Missing old BGG
Avatar
mbmbmbmbmb
I'm working on a geek list (one more in the line of "what games are 'best'/'worst' based on the data in the geek databases). But there are so many ways to look at this data. Please look at the lists below and let me know which best represents your idea of the "best games of 2004".

Thanks.


No.List 1..............List 2..........List 3..........List 4
----------------------------------------------------------------------
01 Downtown:War/Hanoi..War of the Ring.Ticket to Ride..War of the Ring
02 7 Ages..............Power Grid......San Juan........Power Grid
03 War of the Ring.....Memoir '44......Power Grid......Memoir '44
04 ASL Starter Kit.....Goa ............Goa.............Goa
05 Power Grid..........Ticket to Ride..St. Petersburg..7 Ages
06 Struggle of Empires.Maharaja........Memoir '44......Ticket to Ride
07 Memoir '44..........HeroScape.......Hansa...........Maharaja
08 Antiquity...........San Juan........Maharaja........HeroScape
09 Goa.................St. Petersburg..HeroScape.......Doom
10 Doom: Boardgame.....7 Ages..........War of the Ring.San Juan
11 G'burg:Badge/Cour...Hansa...........Oasis...........ASL Starter Kit
12 Ticket to Ride......Doom............Einfach Genial..St. Petersburg
13 Reef Encounter......Einfach Genial..Wings of War....Struggle/Empires
14 Maharaja............Blue Moon.......Blue Moon.......Downtown:War/Hanoi
15 Carc: the City......Axis & Allies...Runebound.......Carc: City
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Chaddyboy
United States
Olathe
Kansas
flag msg tools
admin
designer
Bloooooop.
badge
Bluuuuuurp.
Avatar
mbmbmbmbmb
I agree the most with list #2. Although I haven't played War of the Ring, I like most of the rest of the games. I also wouldn't mind seeing list #3, but I don't agree with it quite as much.

As a player of primarily Euro-games, the wargames on lists 1 and 4 turned me off.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Kane Klenko
United States
Ridgeway
Wisconsin
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
I'm with Chad. Those 2 lists best reflect my thoughts, although I would have Maharaja and Blue Moon as #s 1 & 2.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Seamus Kleissler
United States
Pittsburgh
Pennsylvania
flag msg tools
Avatar
mbmbmbmbmb
i am with Chad, list 2 or 3. Although, I obviously do not own enough games! Need to play more of these!
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Eric Landes
United States
Beaverton
Oregon
flag msg tools
badge
Avatar
mbmbmbmbmb
Either 3 or 4 would be my bet. Downtown got some shilling on here, but reports from people who should know are it's a VERY good game. If you're not restricting yourself to euros, I'd probably pick #4.

#3 would be my second choice.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Dane Peacock
United States
Stansbury Park
Utah
flag msg tools
badge
That tickles
Avatar
mbmbmbmbmb
Gotta go with #4.

This was tough. #4 probably best represents a good mix between my personal favorites and my group's. The underlying decision was whoa, look, camels. caravan sweet! grapecamel
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Ubergeek
United States
Washougal
Washington
flag msg tools
Be Happy in your Game!
badge
I spent 100 Geek Gold and all I got was this lousy overtext message!
Avatar
mbmbmbmbmb
I choose 5
How about a mix-n-match to make a list #5meeple

01 War of the Ring
02 Memoir '44
03 HeroScape
04 Axis & Allies: D-Day
05 Blue Moon
06 Reef Encounter
07 Struggle of Empires
08 7 Ages
09 Wings of War
10 Power Grid
11 St. Petersburg
12 Ticket to Ride
13 Runebound
14 G'burg:Badge/Cour
15 Doom: Boardgame

A lot of the games you chose are the same ones I selected for my own 2004 awards. You have lots of good choices.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Ray
United States
Carpentersville
Illinois
flag msg tools
badge
Avatar
mbmbmbmbmb
#1, which I guess is straight ratings with a minimum number of votes. But don't publish it or it may cause counter-shilling with '1' votes (best to keep it hidden behind a bayesian average ranking which by its nature will always have enough votes to stop counter shilling). Reason being it is so hard to find the small groups of high voters that indicate a great obscure game.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Randy Cox
United States
Clemson
South Carolina
flag msg tools
designer
1024x768 works just fine - Don't Wide the Site!
badge
Missing old BGG
Avatar
mbmbmbmbmb
Thanks for the help. It appears that list 1 is definitely not a valid way to assign best games. That list, by the way, is simply a list of all 2004 releases with at least 30 votes, ranked by absolute, raw average rating here on the 'geek. Obviously, pure average is fairly meaningless, as I had always thought.

The fourth list, which had its adherants, is the Bayesian average list from this here 'geek. Better, by far, than the first list, but not as well received as number 3. In this calculation, all games get some number of extra "average" votes tacked on to their ratings.

The third list is Tim Rafferty's FORM ranking. It's a nifty formula that ranks games along three dimensions--popularity/availability (total number of votes cast), quality (average rating), and reliability/consensus (standard deviation). A case could be made for this being the default rating scale here. And it would be a good case.

The second list, which some supported, is my creation. I call it the "modified Bayesian" list. I add extra votes to games so that each would have enough votes to be among the 200 games with the most votes on the 'geek (that threshhold, by the way, is 399 votes--and that 200th most popular game is Wiz-War). I figure that if a game is reasonably available and popular, it ought to have at least as many votes as Wiz-War, fer cryin' out loud. Games that have average ratings below the norm (average of that list to 200 most popular games) don't get any "Bayesian boost." The rest get enough votes (at average--currently 6.98) to bring them up to 399 total ballots.

What I've done is to create lists of all qualifying games on the geek, broken down by eras. An era is pretty much a year for the past decade. But as you go back in time, some years have to be combined so as to get a decent list of games to work from. I'm planning to post my list of "best games of each year/era" and wanted to know which formula works best.

Thanks for your help.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Joshua Miller
United States
Holland
Michigan
flag msg tools
badge
Avatar
mbmbmbmbmb
Lists #2 and #4 both do a pretty good job. In #1, games without wide circulation (at least among BGG users) are overrepresented - as one would expect of a list that doesn't take into account the number of votes. List #3 is just all out of whack, very bad IMO.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Alex Rockwell
United States
Lynnwood
Washington
flag msg tools
designer
badge
Avatar
mbmbmbmbmb
2 and 4 are very similar, and imo, are the best.

Hey, anything that puts War of the Ring and Power Grid as 1-2 for '04, has got something right.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Geeky McGeekface
United States
Manassas
Virginia
flag msg tools
designer
badge
It's time for baseball, people! Pitchers and catchers report soon and the national pastime is with us again!
Avatar
mbmbmbmbmb
Actually, Randy, I think raw average IS the best way to measure a game's popularity. But you have to wait until it has enough ratings to make the average meaningful. My experience has been that a minimum of 200 votes works well with the Geek rankings. If you look at the 2004 designs and choose the games with the best pure average and at least 200 ratings, you get the following list:

1. War of the Ring
2. Power Grid
3. Memoir '44
4. Goa
5. Ticket to Ride
6. Maharaja
7. Heroscape
8. San Juan
9. St. Petersburg
10. Einfach Genial
11. Axis & Allies
12. Hansa
13. Blue Moon
14. Wings of War
15. Runebound

This list looks an awful lot like yours (list #2), but was a lot easier to develop and has much fewer parameters. Only one of the games even has fewer than 300 ratings (Einfach has 262), so these ratings should be fairly stable.

My list ignores games like 7 Ages and Doom, but these games only have 100 or so ratings. I contend that this is not enough ratings to let early impressions (and rabid fandom) smooth out, particularly against games with at least three times the number of ratings. As I said, a minimum of 200 ratings seems to work pretty well, as can be seen by how close it is to your list.

I prefer this approach to the Bayesian, since it doesn't give an advantage to a game that has, say, 1200 ratings rather than 400. Once a game reaches a certain number of ratings, you're confident that the average rating is accurate, but the Bayesian rating keeps handicapping games with fewer ratings. Besides, the raw average approach is so much simpler.

Really, the only reason that there's any discrepancy between our two lists is that some of the games being considered have been released very recently. If you were to compare your approach to mine for 2003 games, I think you'd find almost perfect correlation.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Daniel Karp
United States
Rockville
Maryland
flag msg tools
admin
Developin' Developin" Developin!!
badge
100 geekgold for OverText, and all I got was this stupid sentence.
Avatar
mbmb
I hate to get into these "how should games be ranked" arguments. (Well, actually, I kind of like it, but it gets me into trouble...) Anyway, I'd like to see a fifth list that is the normal Bayesian average, but with a higher number of "dummy" votes. Bayesian Average, for those who don't know, is:

(A*v + m*C)/(v+m)

Where A is the average vote for the game, v is the number of votes the game received, m is the number of "dummy" votes, and C is the average vote for ALL games. Note that in this case, m is also the minimum number of votes a game has to receive before it is given a Bayesian Average, but there is no reason this must be the case; you could easily use 100 dummy votes and require only 30 votes to have a bayesian average, but of course, in that case, the game's rating would be drawn strongly toward C.

The number of dummy votes is currently set at 30, which was fine back when BGG had significantly fewer users, but now seems a bit low. I would like to see this raised to some higher value--maybe 100 or something. I'd have to see the rankings produced before picking a number. Another alternative (which might work with the first) would be to decrease the value of the dummy votes; currently it is C, the average vote for ALL games, but it needn't be; the purpose of the Bayesian average, as I see it, is to pull down the ranking of games with fewer votes; since the votes on BGG seems to cluster around such a narrow range, adding in "average" votes doesn't have a huge effect. If, instead, the dummy votes had a value of 3, all games would be drawn down, but those with fewer votes would be more seriously affected.

Ultimately, it would be best if the number of dummy votes were indexed in some way to the number of users or the average number of votes on a game or the total number of votes or something, so that the ranking system can grow as teh BGG grows.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Clinton Smith
United States
Port Arthur
Texas
flag msg tools
Avatar
mbmbmbmbmb
Randy, I find it quite interesting that while you consider list #1 flawed because a game with as few as 30 ratings can have an undeserved prominence, you are satisfied that a mere 9 opinions from fellow gamers confirms your personal belief that list #1 is inferior to the other 3 lists.

Nevertheless, I look forward to seeing your upcoming geeklist. It will give me the opportunity to watch someone else take the heat while I relax and enjoy the show.laugh
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Daniel Karp
United States
Rockville
Maryland
flag msg tools
admin
Developin' Developin" Developin!!
badge
100 geekgold for OverText, and all I got was this stupid sentence.
Avatar
mbmb
Quote:
As I said, a minimum of 200 ratings seems to work pretty well, as can be seen by how close it is to your list.


Of course, another way to bring down those games with fewer votes is to kleave them out altogether. And, although I suggested increasing the number of dummy votes while (possibly) leaving the minimum the same, perhaps the reverse would be teh better approach. Unless we see how the lists would come out, it is hard to say.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Randy Cox
United States
Clemson
South Carolina
flag msg tools
designer
1024x768 works just fine - Don't Wide the Site!
badge
Missing old BGG
Avatar
mbmbmbmbmb
Larry,

I, too, think raw average is ok, with sufficient votes. But I think "sufficient" is more like 400, not 200. That's why I add average votes only to bring a game's tally up to 400 (-ish). If it already has 400, I don't change it's average.

But you can't just eliminate those new games that haven't yet received enough votes to qualify. You have to guesstimate somehow, at least in my book.

And just eliminating those without 200 votes will still miss many in my prior-year lists (Bridge, Twixt, Focus...and that's just from the oldest list).
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Randy Cox
United States
Clemson
South Carolina
flag msg tools
designer
1024x768 works just fine - Don't Wide the Site!
badge
Missing old BGG
Avatar
mbmbmbmbmb
Daniel,

Using your "Bayesian 200" calculation with a standard entry of 3, rather than the average rating (which I agree whole-hardedly with, because I may not be rating the game because I'd never play it in a million years), here's the Top 15 of 2004:

1. Ticket to Ride
2. Memoir '44
3. Power Grid
4. San Juan
5. Goa
6. War of the Ring
7. St. Petersburg
8. HeroScape
9. Maharaja
10. Hansa
11. Runebound
12. Betrayal...Hill
13. Blue Moon
14. Axis & Allies
15. Wings of War

To me, it's an improvement over current Bayesian, but not by a lot. I still like adding fewer (or no) dummy votes to games that already have plenty of ballots cast.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Daniel Karp
United States
Rockville
Maryland
flag msg tools
admin
Developin' Developin" Developin!!
badge
100 geekgold for OverText, and all I got was this stupid sentence.
Avatar
mbmb
Quote:
Daniel,

Using your "Bayesian 200" calculation with a standard entry of 3, rather than the average rating (which I agree whole-hardedly with, because I may not be rating the game because I'd never play it in a million years), here's the Top 15 of 2004:


Interesting... I would say, though, that both significantly increasing the standard entry and significantly decreasing the standard vote might be a bit extreme. Now you are REALLY dragging down the games with fewer votes. As for adding few or no dummy votes to those games with plenty of votes, it shouldn't matter much--the more votes a game gets, the less significant those dummy votes will be anyway--that is, unless you add 200 votes of 3.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Ben Vögel
United States
Golden Valley
Minnesota
flag msg tools
badge
I've been on this site for 15 years now, and I'm far from sick of games, but I think I prefer a better balance of favorites to new games. I'm also tired of playing 4+ hr multiplayer games, but I'll still happily play really long games 2 player.
Avatar
mbmbmbmbmb
If you took a poll of the entire population, not just gamers, list #3 would most accurately reflect the general population's experiences with these games. That list shows the games with which gamers have had good luck introducing to non-gamers. That list is the list that has the most usefulness in guiding folks with little exposure to good board games into games they will enjoy.

If the goal is to satisfy all the folks here at BGG with a definitve list, well that just isn't going to happen. We already know quite a bit about games. We know what themes, mechanics, complexities, and other atributes we find appealing in games. We know what we think are the "best" games, and no rating list alone is talking us out of it, no way, no how.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Darren M
Canada
Fort Vermilion
AB
flag msg tools
badge
Avatar
mb
My own take on the best of 2004:
Using my own ranking method...which combines and takes into account average ratings,bayesian averages and standard deviations and gives equal emphasis to all 3...I get the following list of the top 25 games for 2004:

1 Downtown: Air War Over Hanoi, 1965-1972
2 Power Grid
3 7 Ages
4 Goa
5 Ticket to Ride
6 Memoir '44
7 San Juan
8 Maharaja: Palace Building in India
9 Oltremare - Merchants of Venice
10 Carcassonne die Stadt
11 Reef Encounter
12 Struggle of Empires
13 War of the Ring
14 Doom: The Boardgame
15 St. Petersburg
16 Advanced Squad Leader (ASL) Starter Kit #1
17 Hansa
18 HeroScape
19 Antiquity
20 Einfach Genial
21 Gettysburg: Badges of Courage
22 Sword of Rome, The: Conquest of Italy, 362-272 BC
23 Ys
24 Das Zepter von Zavandor
25 Lost Valley



I like to think of this method as good for showing which games are the most suitable for a broad range of gamers and gaming groups. There are games in there that are still niche games...but they are great in their niche and I don't like methods that exclude games simply because they haven't been rated by a lot of individuals or aren't widely known yet.

It's then easy enough to simply say ...well I don't like wargames and you just knock out all wargames in the list...or if you dislike Eurogames...it's just as easy to knock those off instead and that way you don't have to come up with artificial methods of excluding games that are great in their genres even though they have no personal appeal.




 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Deleted User
msg tools
badge
Avatar
#3

Is there one WITHOUT War of the Ring on it? I'd vote for that one.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Darren M
Canada
Fort Vermilion
AB
flag msg tools
badge
Avatar
mb
Some games have wider variance of opinions than others.
That's one of the reasons I put more emphasis on Standard deviation...some games (like War Of The Ring) have high ratings but also have pretty high SD levels which shows there is a lot of spread in the opinions of the game which for me...makes the game a bit less appealing as there is a higher chance that a gaming group will disagree on how interesting and enjoyable the game is.

I personally like games that are a little more "middle of the road"... meaning that maybe the ratings aren't quite as high for them...but there isn't as high of a "hate it or love it" feel to the game as well.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Randy Cox
United States
Clemson
South Carolina
flag msg tools
designer
1024x768 works just fine - Don't Wide the Site!
badge
Missing old BGG
Avatar
mbmbmbmbmb
Quote:
you don't have to come up with artificial methods of excluding games that are great in their genres even though they have no personal appeal.


Well, Darren, we simply disagree on this one. For me, I don't want teensy-tiny niche games that appeal only to a handful of people appearing on the broad-spectrum mass-appeal Best of the Year lists. Maybe it's just me, but I don't consider Coffee Milk to be the greatest drink ever (though it's the State Drink of Rhode Island). Local (I could say "provencial" to talk down) tastes just don't mean much when looking at the global picture. Likewise, a great wargame with 42 strong adherants isn't important when I'm trying to come up with an across-the-board, appeal-to-as-many-BGGers-as-possible list of best games.

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Joshua Miller
United States
Holland
Michigan
flag msg tools
badge
Avatar
mbmbmbmbmb
Randy, I think I can offer a simple method that gives what looks to me like a more accurate list than any we've discussed so far. I'm just using a Bayesian average with a much larger penalty to less frequently-rated games: 100 baseline votes with a rating of '5.'

Here are the results. I'll go all the way to #25 so you can see what you think. The big gap after game #9 is interesting, and looks right to me.

1. Power Grid (7.817)
2. War of the Ring (7.795)
3. Ticket to Ride (7.756)
4. Goa (7.676)
5. Memoir '44 (7.676)
6. San Juan (7.486)
7. Saint Petersburg (7.388)
8. Maharaja (7.371)
9. HeroScape (7.329)
10. Hansa (6.981)
11. Blue Moon (6.846)
12. 7 Ages (6.841)
13. Doom: The Boardgame (6.831)
14. Einfach Genial (6.820)
15. Axis & Allies (6.816)
16. Runebound (6.775)
17. Wings of War: Famous Aces (6.754)
18. Betrayal at House on the Hill (6.620)
19. ASL Starter Kit #1 (6.579)
20. Carcassonne: Die Stadt (6.567)
21. OltreMare (6.496)
22. Reef Encounter (6.470)
23. Oasis (6.446)
24. Struggle of Empires (6.437)
25. Axis & Allies: D-Day (6.3591)

The top ten all have at least 470 votes. Games with around 100 votes can still do well, but they'll fall down the list. 7 Ages has 108 votes, but with an incredibly high average, and finishes #12. In the 16-25 range, you start to see a mix of games with many "solid but unspectacular" ratings (Runebound, Oasis, Betrayal, Wings of War), and some of the games with fewer ratings but a high average (ASL Kit #1, Carcassone Stadt, OltreMare, Reef Encounter, Struggle of Empires).

To me this list does a great job of reflecting what I view as the consensus surrounding the games of 2004. What do you think?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Darren M
Canada
Fort Vermilion
AB
flag msg tools
badge
Avatar
mb
less votes does not always mean a lesser game.
My view is that essentially the number of votes a game has is only an indication of how widely played a game is...and only in general terms. Settlers,Carcassonne and Puerto Rico are probably the 3 most widely known Eurogames hence they have the most votes on BGG. Monopoly is surely one of the most widely known and played modern games worldwide and even it has far fewer votes than many games here on the geek...as do Go and Chess for that matter.

There is certainly a slant here towards Euro/German games which is perfectly understandable as these are the types of games that the boardgame hobby has evolved towards in recent years and this is the genre that has produced some of the most interesting designs lately.

I think average rating of a game is an absolute and unbiased measure of a game's worth as an "entertainment vehicle". As long as we assume that all ratings are valid and equal and express each individuals true opinion on the games they play...then the average rating is truly a valid view of a games "worth".

There of course will be shills and vendetta type votes on BGG and these do effect the games with a lower number of ratings which is why the Bayesian average is useful as well...to penalise games with fewer votes until they get more opinions and the shills and vendetta votes have a chance to be balanced by unbiased votes. Common knowledge in statistics is that a valid sample of a populaton of votes is around 25-30 before any real validity can be given to the results...so yes I do think a minimum level of approx 30 votes needs to be cast before we really have a decent impression of a games value to the BGG community.

The more votes of course...the more certain we can be that the results are truly representative of the BGG community. There is a point though where there is diminishing returns...we don't really need 200 or 500 or 1000 votes to get a good idea of how well a game is being recieved. If a game has a 9 average rating after 30 votes...it's pretty safe to say it's going to have a lower rating as time goes on...as almost every game drops it's average rating as it gets more votes...simply because people get bored with the game and it loses it's newness and novelty factor. But that game with a 9 rating is probably going to be pretty highly rated after 100,500 and 1000 votes as well.

If that game never recieves 100 or 500 or 1000 votes that doesn't mean that it isn't a great game...it just means that it's a niche game that isn't widely played or it may be a limited edition...etc.

Puerto Rico started with 1 vote on BGG...and it was a highly regarded game with 30 votes...and at 100 votes and at 500 votes and at 3000 votes...and yes it's rating dropped as time went on but it still is very highly thought of.

I can see now that new games like 7 Ages,Struggle Of Empires and even the up and coming Revolution: The Dutch Revolt 1568-1648 are going to be games that are potentially top 25 type games on BGG...and I think their average rating is more representative of their ranking than their bayesian at the moment...BUT an average of the Bayesian and the average rating is probably an even better estimate of where they will end up...as the average ratings will drop in time and the bayesian average will creep up as the number of votes cast will rise.

The last job is then to separate the games into genres...because there simply is really no valid way to compare games like Puerto Rico,Warmachine and Paths of Glory to each other. They are 3 totally different beasts which are all very highly regarded in their genres...but comparing them head to head is certainly apples and oranges.

Puerto Rico may be rated higher than WM or PoG but that means nothing if I'm a mini or wargamer.

BGG is a boardgamer's site not just a Eurogamer's site so having the genres of minis,wargames,party games...etc is great and open up doors for us to learn about games that we would have no idea about otherwise.

So yes...in the end any ranking method anyone uses is valid as long as it suits the criteria they choose to look at when they want to pick games they know will suit their gaming groups.



 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
1 , 2  Next »   | 
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.