The h-index is defined as the maximum value of h such that the given author/journal has published h papers that have each been cited at least h times.  So, applying it to board games, the h-index is defined as the maximum value of h such that a person has played h different games at least h times.  Many gamers (myself included) have calculated this value for themselves; a few (again, including me) have even aimed to reach an h-index of at least 50 before they turn 50.

But thinking more about various h-index lists I’ve seen – there’s a lot of overlap between them.  And so I thought it would be interesting to see what commonality there is between various Opinionated Gamers writers.  

In the end, 16 writers provided their h-index lists; the h-indices themselves range from 10 through 54, and 303 different games were listed.  These lists – and the overlaps – were fairly interesting in and of themselves; there were plenty of shorter games listed, but a number of folks had multi-hour games included as well; 1870 and 2038 were each included once, for example, and by different writers.

But looking at the overlap, I realized that I wanted another piece of data – how much, in general, are we playing these games?  And unlike h-index, which requires reasonably precise data which not all of the writers collect, here I wanted to allow for approximation.  And, correspondingly, I decided to use median plays – the number of plays half of the writers reported fewer than, and half reported more than.  I collected some data for all of the games on two or more h-index lists, but since there were so many such games – 100, not including “Unpublished Prototype” – I asked folks to focus on those 48 games on at least three such lists.

19 of the writers provided their data, whether precise, approximate, or a mix of the two.  And the data was enlightening – by restricting to games that _some_ of us had played a lot, we ended up with a list of games that nearly _all_ of us had experienced.  Every single respondent had played at least 40 of the 48 games, and three had played all 48.  And as you’ll see, only one of the games was truly a niche selection, with more folks not having played it than had.

So which games do we play so often that they frequently end up on h-index lists?

Carcassone – median of 50 plays, on 11 h-index lists: This game showed up on the most h-index lists, and tied for the highest median number of plays.  This really speaks to the universal nature of the game – and reinforces the idea I had, when collecting this information, that the overlaps of h-indices would provide not so much a list of “games you should play” and “games we all have played” – not inherently a greatest hits, but a common ground.

Ticket to Ride – median of 50 plays, on 6 h-index lists 

Can’t Stop – median of 44 plays, on 7 h-index lists

Dominion – median of 38 plays, on 9 h-index lists

Puerto Rico – median of 30 plays, on 8 h-index lists: The surprise to me here was a personal one – that I’d play the games the most times of those who responded.  

Codenames – median of 28 plays, on 7 h-index lists

Lost Cities – median of 28 plays, on 7 h-index lists

Diamant – median of 25 plays, on 5 h-index lists

Crokinole – median of 23 plays, on 6 h-index lists, not played by one respondent

For Sale – median of 23 plays, on 5 h-index lists

Die Siedler von Catan / Settlers of Catan / Catan – median of 21 plays, on 3 h-index lists

Tichu – median of 20 plays, on 4 h-index lists

Cribbage – median of 20 plays, on 3 h-index lists, not played by five respondents: Counting down the median values, this was the first game never to have been played by a significant number of those who provided data.  This is also the oldest game on this list.

Loopin’ Louie – median of 19 plays, on 7 h-index lists, not played by one respondent

Geschenkt / No Thanks – median of 18.5 plays, on 6 h-index lists, not played by two respondents

7 Wonders – median of 17 plays, on 5 h-index lists

TransAmerica – median of 17 plays, on 3 h-index lists, not played by two respondents

Agricola – median of 16 plays, on 6 h-index lists

Liar’s Dice – median of 15 plays, on 7 h-index lists, not played by one respondent

Roll for the Galaxy – median of 14 plays, on 3 h-index lists

Thurn & Taxis – median of 14 plays, on 3 h-index lists

Saint Petersburg – median of 12 plays, on 7 h-index lists

Race for the Galaxy – median of 12 plays, on 5 h-index lists

Bohnanza – median of 12 plays, on 4 h-index lists

Azul – median of 12 plays, on 3 h-index lists: Much to my surprise, this was the only game among these 48 which I had played the least of anyone who responded.

Web of Power – median of 12 plays, on 3 h-index lists, not played by one respondent

Age of Steam – median of 11 plays, on 3 h-index lists, not played by two respondents

The Princes of Florence – median of 11 plays, on 3 h-index lists

Ra – median of 10 plays, on 5 h-index lists, not played by one respondent

San Juan – median of 10 plays, on 5 h-index lists

Roll Through the Ages: The Bronze Age – median of 10 plays, on 3 h-index lists, not played by one respondent

The Mind – median of 10 plays, on 3 h-index lists

Alhambra – median of 9 plays, on 3 h-index lists

Einfach Genial / Ingenious – median of 9 plays, on 3 h-index lists

Pandemic – median of 8 plays, on 3 h-index lists

Terraforming Mars – median of 8 plays, on 3 h-index lists, not played by one respondent

Euphrat & Tigris – median of 7 plays, on 4 h-index lists, not played by one respondent

Die Crew – median of 6 plays, on 4 h-index lists, not played by one respondent

Traumfabrik / Dream Factory – median of 5 plays, on 4 h-index lists, not played by three respondents

Zirkus Flohcati / Circus Flohcati – median of 5 plays, on 4 h-index lists, not played by four respondents

Mamma Mia – median of 5 plays, on 3 h-index lists, not played by five respondents

Ave Caesar – median of 4 plays, on 3 h-index lists, not played by five respondents

Jump Drive – median of 4 plays, on 3 h-index lists, not played by five respondents

Fairy Tale – median of 3 plays, on 3 h-index lists, not played by three respondents

Innovation – median of 3 plays, on 3 h-index lists, not played by one respondent

Mystery Rummy 1: Jack the Ripper – median of 2 plays, on 3 h-index lists, not played by three respondents

One Night Ultimate Werewolf – median of 2 plays, on 3 h-index lists, not played by six respondents

Fast Food Franchise – median of 0 plays, on 3 h-index lists, not played by ten respondents: Of all of the games listed by three or more folks on their h-index, this was clearly the most niche.  The three folks who included it have all played it 100 or more times; everyone else who has played it has played fewer than 50 times combined.

Among games on 2 H-index lists, the games with the highest median number of plays were Hearts, 6 Nimmt!, Poker, Scrabble, Connect Four, Power Grid, Splendor, Amun-Re, The Castles of Burgundy, Durch die Wüste / Through the Desert, Terra Mystica, Linko!, and Pickomino.

So – how many of the top 48 games have you played?  And which are part of your h-index?

Comments from other Opinionated Gamers

Dale Y- so i’m not a stats guy. I honestly stopped tracking most of my game plays back in the Naughties.  So, I was one of the guys here who simply estimated their plays.  I kept pretty good records of games that I developed, so I was sure of my 1200+ games of Dominion and 200+ games of Agricola.  For the rest, I had to estimate to the nearest 10 or 20.  My list skews more to the older games because man, back in 2000-2005, I only got access to 5 to 10 new games a year, and we played the heck out of all of them.    Of the top 12 games, I have played all of them many many times, and would be happy to play any of them again except for Puerto Rico.  I think that those games give a pretty interesting take on what might be a “universal” game collection.   I also thought it was interesting to see how many games had been played by the large percentage of the responders.  I’ll leave the rest of the stat dissection to the gearheads on the list.

Andrea “Liga” Ligabue: I have not sent my lists since I’m tracking “more than sporadically” games played just starting from 2020. I’m also using games a lot as an activity in schools and I’m sure I have teached more than 1000 kids playing Jamaica, Super Farmer, Carcassonne, Ticket to Ride, Stone Age and Kaleidos and more than 500 kids Stone Age Jr and Viva Topo. 

I’m quite sure, as Dale says, that back in the 90’ and in the beginning of this century, with less interesting games, I have played countless games like RoboRally and Battletech (1-2 games every week … every!!!). 

Of course it is easiest to play often short games than long games but I’m tracking all the results of my Terraforming Mars plays and I’m sure to have played more than 40 games from summer 2019 to now.

Talia Rosen: Unlike Dale and Liga, I’m obsessed with game stats and I track them diligently.  This was probably evident from my 10,000 Plays and Counting post last year.  That being said, I’d never heard of an “h-index” before Joe told us about it a few weeks back.  Well, maybe I heard about it in my 1999 statistics class, but if so, then I completely forgot.  I had fun calculating that my own h-index is 42, and seeing what games just made or missed the cut.  I was happy to see Root just make the cut, having fallen for it head over heels.  I was surprised to see Age of Steam and Agricola make the cut, having not played either in a while, but I guess I used to play both a lot.  It was a shame to see Notre Dame and Antiquity both barely miss the cut because I really enjoy both of those a ton, but I suppose it’s just more incentive to play them.  As it turns out, I’ve played all of the games mentioned in the article above, except Circus Flohcati, Jump Drive, Mystery Rummy, and Fast Food Franchise.  I’d never even heard of Jump Drive!  I guess this is motivation to seek out and try those four games, given that they’re on so many other h-indices.  This was a fun exercise for a stat that I’d never heard of, and I encourage everyone else to try it out for themselves and report back below.