2008 Failure Rate of Roguelike Games

By Jeff Lait for rec.games.roguelike.development

It is now time for my fifth annual analysis of roguelike game development. I shall first present some dubious statistics and then you shall complain that they don’t accurately reflect roguelike development.

To find the previous four studies, search for Failure inside this newsgroup.

The data for this comes from:http://thelist.roguelikedevelopment.org/ which I have been maintaining.

First, the meaningless bargraph.

1     #
1     #
1    ##
1 #  ##
1 #  ##
1 #  ##          #
1 #  ##          #                       #
1 #  ##          #           #           #
1 # ####         ##          #           #
1 ###### #      ###          #           #
1 ###### #  #   ### #        #           #
1 ###### #  #  #### # #      ##     #  # #
1 ###### ############### ##  ##  # ## ####
1 ###################### ####### #### ######
1 000000000111111111122222222223333333333444>+49
1 123456789012345678901234567890123456789012>

This tracks the number of roguelikes by last release date. The first column has a # for every roguelike released in the last month. I have omitted the last column which would have all the roguelikes over 42 months old or without known release dates. There are now 49 such roguelikes being tracked.

The peaks at 4-5, 16, 28, and 40 month marks are due to the 7DRL challenges. Note that a large number of 7DRL entries occurred pre-challenge this year resulting in the spread of the 7DRL effect.

Next, we will look at the cumulative totals for the last year.

Numbers: (July, 2008)
Month   #   Total  Percent
1      11      11       6%
2       5      16       8%
3       6      22      11%
4      12      34      17%
5      14      48      24%
6       6      54      27%
7       1      55      28%
8       5      60      30%
9       2      62      31%
10      2      64      32%
11      4      68      35%
12      2      70      36%
Rest  127     197     100%

Copying from the last three year’s reports and reconstructing the 2004 numbers:

Numbers: (July, 2007)
Month   #   Total  Percent
1      10      10       6%
2       6      16      10%
3       9      25      15%
4      11      36      22%
5       9      45      28%
6       5      50      31%
7       5      55      34%
8       3      58      36%
9       3      61      37%
10      2      63      39%
11      1      64      39%
12      2      66      40%
Rest   97     163     100%

Numbers: (July, 2006)
Month    #    Total   Percent
1        9        9       7%
2        3       12       9%
3        3       15      12%
4       11       26      20%
5        5       31      24%
6        1       32      25%
7        2       34      26%
8        3       37      29%
9        1       38      29%
10       3       41      32%
11       4       45      35%
12       2       47      36%
Rest    81      128     100%

Numbers: (July, 2005)
Month  #  Total  Percent
1     15     15      15%
2      3     18      17%
3     10     28      27%
4     12     40      39%
5      2     42      42%
6      1     43      42%
7      5     48      47%
8      2     50      49%
9      3     53      51%
10     2     55      53%
11     3     58      56%
12     2     60      58%
Rest  43    103     100%

Numbers: (July, 2004)
Month  #  Total  Percent
1      6      6      10%
2      5     11      19%
3      2     13      22%
4      3     16      27%
5      0     16      27%
6      0     16      27%
7      4     20      34%
8      0     20      34%
9      0     20      34%
10     1     21      36%
11     2     23      39%
12     2     25      42%
Rest  24     59     100%

The original metric I measured, Percent Actively Developing Roguelikes, is clearly becoming meaningless as the natural churn of roguelike development will send this number to zero. I am still keeping it in the table for completeness. Interestingly, it is still bouncing around the 40% mark – showing roguelikes are still in a growing curve from the year-zero when I built the original data.

More interesting is the absolute number of touched roguelikes. 2006 seems to have been an anomaly as we’ve continued to see growth in this area with 70 roguelikes updated in the last year.

This chart shows the number roguelikes touched in the last 6 months, 12 months, and the percentage the twelfth month number comprises of the total number of roguelikes being tracked.

Year    6     12     %     Total   New
2004   16     25   27%        59     -
2005   43     60   42%       103   +44
2006   32     47   36%       128   +25
2007   50     66   40%       163   +35
2008   54     70   36%       197   +34

I think after five years we can start to say something meaningful about these trends. 2008 saw similar patterns to 2007 – a lot of new roguelikes but also a lot of old timers resurfacing to the top.

The absolute numbers are equally impressive – 70 projects saw another point release in the last year. Of those, an astounding 54 were last updated in the last six months. Roguelike creation, as measured by roguelikes making it to this list, has tracked consistently at three roguelikes per month for the last five years!

—
Jeff Lait, July 2, 2008