A Year End Analysis of Slam Jams – Part 2

I’m back with the second part of attempting to look at Data from the first four Slam Jams. This time, I’m going to attempt to look more at cards and how popular/successful they have been.  Before I do that though, I just wanted to show one graph I left out of the last analysis, showing the popularity of Phoenixborn:

pbpopularity

A reminder that October was the same weekend as a large amount of APC qualifiers, and so the turnout was low.  In general, people have played a pretty diverse set of Phoenixborn – which is great to see.  We still haven’t seen any Maeoni, and many other good PBs are underrepresented in the field (e.g. Coal).

Let’s get to the cards.

topcardsbydeckcount topcardsbycopies

These are the most popular cards, in the field, by two different counts.  The first graph shows the total number of decks playing a card (out of a possible 48), and the second graph shows the number of total copies. Some things that stand out to me:

  • It’s pretty surprising to me that Frostback Bear is so popular it makes it into the copy count despite often being seen as a 1-of.
  • No charm cards make the top 10 of either count.
  • Ice Trap is extremely popular (is this why Aradel has been undersuccessful?)
  • By count, one card of each non-charm takes the top 3 spots, and each of them are fairly close.
  • I’m not sure Molten Gold is the most powerful nature card, but its ubiquity is definitely shown here with how popular it is.

screen-shot-2016-12-28-at-11-41-49-am screen-shot-2016-12-28-at-11-39-23-am

These graphs (which exclude October), attempt to show which cards are overperforming or underperforming by looking at what % of the card is seen in the Top 4 of tournaments versus the entire field. For instance, if a tournament had 16 decks and 8 copies of Summon Gilder, we would expect exactly 2 copies of Summon Gilder in the top 4.  This would be seen as 100% in the graph.  More than 2 copies would be > 100%, and less would be, ofcourse, less.

Some observations:

  • Shifting Mist is by far the most overly successful card – seeming to do great by a huge portion of people playing it.
  • Reaction spells make up 3/5 of the undersuccessful cards.
  • There are some surprises here.  I think part of them can be explained by flaws in the methodology; this graph tends to overly reward cards favored by strong players and punish cards favored by weaker players – regardless of how good those cards actually are.
  • Charm cards make up 1.5 of the bottom 5

screen-shot-2016-12-28-at-12-56-33-pm

Here is representation for the outlier charm cards (no limits on number of copies). I think there’s a couple things to be noted here:

  • Sympathy Pain looks bad.  I think a large part of this can be seen because of the meta and decks playing it.  2/3 of all decks running Sympathy Pain ran 3 or less charm dice, and only 1 deck ran more than 4.  In a world where illusion dice not only exist but are popular (especially amongst successful players), this is just too easily punished.
  • Three-eyed owl, despite being used by successful decks (top 2 of last slam jam, winner of October, top 2 of Gencon), looks the worst.  I think this, again, highlights the flaw in the methodology. A card may look worse than it is if it is used often in incorrect ways.
  • Charm may just be tricky to play. The poor looking cards are still played by successful decks, and the good looking cards are used almost exclusively by successful decks.cardpowerrankings

cardnonweightednormalized

One last look at card representation.  These graphs (both of which require a card to appear in a minimum of 5 decks) attempt to look at card success by assigning points to their placement when they are present in a deck.  A winning deck will earn 10 points, 2nd place 5 points, etc.

The top graph attempts to normalize this by dividing the number of decks seen playing the card.  It will attempt to highlight cards that are indeed quite powerful, but will overly reward scarcely played cards and punish cards that are played by nearly everyone (like Molten Gold).  The second graph does not normalize in this way.  It does, however, have a maximum score of 1.0.  It’s flaw is that it rewards cards commonly played – if every single player used Molten Gold, it would automatically receive a 1.0

Some observations:

  • The top graph pulls out some cards that we know to be successful, not yet highlighted by other metrics.  Finally, Fear, Regress, and Three-Eyed Owl all take top spots and look very good.
  • Vanish looks surprisingly dominant – this is perhaps because of the overwhelming popularity of Molten Gold.
  • Butterfly Monks look great on both graphs, which is a high indicator that they are probably just a stone-cold bomb.

screen-shot-2016-12-28-at-10-18-08-am

These are popular conjurations over time (excluding October, due to its low showing).  It’s interesting to see Shadow Spirits slow rise to the top, and Frostback Bears slow decline.  The bears aren’t just declining in popularity, however…

screen-shot-2016-12-27-at-9-59-31-pm

This is the % of copies of Frostback Bear seen in the top 4 over time.  If 1 copy was played in the field, and 1 was in the top 4, this would be 1.0.  It’s been declining – which is tangential to it’s declining popularity.  Are people truly learning to handle the bears, or are good players simply forgoing them for other conjurations?

Final thoughts:

  • People need to play more charm cards, and play the dice to support them. Beast Tamer has been seen a shocking 0 times in Slam Jam decks (it’s probably the best card to have that stat).  If you are going to play them, though, you have to have the dice to back them up.
  • The Focus X pair of Shadow Spirit and Butterfly Monk are becoming more and more prevalent, and their best friend Shifting Mist is seeing success because of it. If you’re playing, I would consider how your deck handles both of these.
  • Hammer Knight is good and underplayed. Bring him back!
  • Cut the Strings is the most successful card if you set no minimums.  It’s only appeared in 2 decks, and those decks took 1st/2nd.

Leave a Reply

Your email address will not be published. Required fields are marked *