I’ve been spending a lot of skull sweat the last couple weeks, working out a revision to our points ladder (and related points distribution between assignments).  The initial ladder I developed worked OK, but there are several things I noticed as we’ve played through the game this year that I want to adjust to improve the game flow for the next revision:

  • I knew enough of game design to make the first few levels quite easy to achieve.  But I didn’t pay enough attention to the progression past the first 3 levels.
  • I didn’t even consider the spread between semesters, which my students were quick to notice as we began, the first couple of days.
  • The later levels should be harder to achieve than the earlier ones.
  • I stopped my ladder at a point equivalent to an A+…I want at least a couple levels higher than that, for anyone who gets there before the end of the term.
  • I initially built the ladder to reflect a 1000 point distribution, to easily map onto the A-F scale used at school.  <shudder> a bad idea.  (more on this in another post)

My initial ladder looked something like this:

Level XP Level XP
N00blet 1 20 L33t 11 670
N00blet 2 60 GoSu 12 700
N00b 3 110 GoSu 13 730
N00b 4 170 GoSu 14 770
N00b 5 240 Beast 15 800
N00bLord 6 320 Beast 16 830
N00bLord 7 410 Beast 17 870
N00bLord 8 510 Boss 18 900
L33t 9 600 Boss 19 930
L33t 10 630 Grandmaster 20 1000

If you’re paying more attention than I did at first, you’ve already noticed that the point spread between individual levels varies radically, from 20 to 100 points.  By itself, that isn’t necessarily a bad thing (the spread should increase as the levels do).  But the randomness with which I changed it is a problem.  Having to earn 100 points to go from level 7 to level 8, but then only needing 90 for the next level and only 30 for the level after that?  Bad game mechanics.

So, as the course progressed, I noticed these things (sometimes with the help of my students) and have been modifying my ladder to reflect the more sophisticated understanding of the mechanic that I’ve developed.  😉  I’ve made some changes to the current ladder, but I didn’t want to completely wreck the current game, so not all the changes I want have gone into this year’s iteration.

Here’s the plan for next year:

Level XP Spread Level XP Spread
1 N00blet 15 15 11 L33t 500 75
2 N00blet 40 25 12 GoSu 575 75
3 N00b 75 35 13 GoSu 650 75
4 N00b 110 35 14 GoSu 730 80
5 N00b 150 40 15 Beast 810 80
6 N00bLord 200 50 16 Beast 890 80
7 N00bLord 250 50 17 Beast 1010 120
8 N00bLord 300 50 18 Boss 1130 120
9 L33t 350 50 19 Boss 1250 120
10 L33t 425 75 20 Grandmaster 1400 150

The Spread column indicates the difference between the previous level and the listed one.  This progression is much smoother, though I’m still not quite happy with it.  I think the distribution should jump more between levels 13 and 14, for one thing.  But in conjunction with this progression, I have to assign points to the various quests and craftings (assignments) that the students complete.  Coordinating this with those has been a bit of a struggle.  I think this will be a much better flow though, because

  • Jumps between levels are low enough at the beginning to allow for rapid progression
  • Jumps quickly become more difficult after the first few levels, allowing for more challenge to keep players engaged
  • The ladder extends well beyond the credit students achieve for the basic quests and craftings; to reach levels 18-20 they’ll have to do a substantial number of optional quests and to reach level 20 requires very careful planning and truly excellent execution
  • The difference between semesters is still substantial, but for the game mechanics to keep the challenge high as students progress, that’s required.  I’m debating whether to tie the grade to the points earned next year.  Right now, I’m leaning toward figuring out some indirect way to assign grades, because I haven’t been happy with the direct link results at all.

Do you use a level ladder with points?  How has yours worked?  Any lessons to share?