In their Sunday afternoon panel discussion Ryan and Chelsea Neil expressed a desire for creating professional standards in bonsai and for consistency and transparency in judging of bonsai exhibits. To that end, the judges at The Artisans Cup were separated during judging so they would not influence each other and they were informed in advance that the raw scores for all judges would be published on The Artisans Cup blog.
Bay Island Bonsai members and Intensive Students who study with Boon Manakitivipart are all familiar and practiced in scoring trees on a 30 point scale. The purpose of the exercise is to teach a method for evaluating each aspect of a tree against other trees of the same species or of similar character. Boon’s rubric emphasizes the importance of the trunk by awarding 1-10 points, the branching is scored 1-5, the nebari receives 1-5 points and the container merits a similar 1-5 score. The last five potential points are scored on the overall impression, giving a judge the chance to score higher or lower for trees that may be unusual but still merit consideration.
The rubric for the judges as published on The Artisans Cup blog uses a single score 0-60. While the instructions are extensive for aspects of the tree and display composition that should be evaluated, no specific guidelines for each aspect are detailed numerically. Instead, consistency in judging across all trees is emphasized. While we could speculate that more instructions were provided on the actual scoring sheet, it is safe to assume for this exercise that this is the extent of the directions the judges received.
The published results give bonsai enthusiasts an unprecedented opportunity to use statistics as a tool for learning. Herewith, a rudimentary analysis of some trees from the event which I hope will give some insight into what five highly-respected judges believe qualifies as a great bonsai at one of America’s top bonsai shows.
The average score for all trees accounting for the discarded low and high scores in the rubric was 43 with a standard deviation of 4. Predictably, two standard deviations above mean (51) gives us the two winners which went to a non-statistical tiebreaker.
High Scoring Trees
Let’s take a look at the set of trees that were one standard deviation above the mean (47) or higher; there were 14 trees in this category. Based on the information in the program, here are some statistics:
13 of 14 were conifers, the single non-conifer was a Buttonwood.
Correction: All 14 of the top 14 were conifers. The Buttonwood scored 46.
11 of the 14 highest scores went to domestically collected wild trees. 3 trees were Japanese imports, 1 origin not listed.
14 of 14 highest scores went to trees that were at least 18″ across or 18″ tall.
7 of the 14 were a juniper species.
3 of the 14 were white pines.
2 of the 14 were Redwoods.
1 was a Mountain Hemlock.
1 was a Douglas Fir.
Low Scoring Trees
11 single trees and 3 multi-tree displays scored one standard deviation below the mean or lower. Here are some statistics in this set:
10 of 14 were nursery stock or field grown.
2 of 14 were domestically collected wild trees.
2 of 14 origin unspecified.
There is no clear size trend among low scoring trees.
2 of the 14 were a juniper species.
8 of the 11 single trees were conifers.
3 of the 11 single trees were broadleaf.
So which tree was the favorite of each judge based on raw scores? Three are shown above, here are all of them:
Peter Warren awarded top marks to Eric Schikowski’s collected Mountain Hemlock.
David Degroot awarded the high score to two trees: The large Japanese Black Pine from the Pacific Bonsai Museum and Jim Gremel’s field-grown ‘Kishu’ Chinese juniper.
Colin Lewis gave top prize to Doug Paul’s collected Rocky Mountain Douglas Fir.
Walter Pall was most impressed by Bob Shimon’s collected coast redwood.
And Boon marked
John Kirby’s collected California Juniper as most deserving Correction: Boon picked Eric Schikowski’s collected mountain hemlock.
As expected from the Rubric, each judges’ high score was not counted in the final average for their favorite tree.
While statistics may tell us a few things about this show they may also be misleading with a show that was only 71 displays deep. The trends on the high scoring trees are clear with a near-unanimity for collected conifers. 68% of the entire show was made up of conifers, 92% of the high scoring trees were conifers. The trend in sizing is also pretty clear – medium to larger trees scored higher.
Based on these results, it’s safe to say that if you want to impress these bonsai judges, you’d better have large collected conifers. Indeed, if we consider the basic criteria for judging any bonsai, it is logical that the collected conifers scored highly. The collected trees have trunks that are readily available in larger sizes and that already have a lot of character.
What is more difficult to understand is why some of the best broadleaf trees in the exhibit did not score higher. There were certainly a number of very high quality entries. Did the collected conifers score higher because they inherently possess more bonsai character? Or was it because the deciduous in the show were not as well-developed in this case?
Perhaps this means that we all have a lot of work to do when it comes to creating deciduous and broadleaf trees that will be of equal or higher quality to the collected conifers of the Western US. Many of these conifers had been in training for fewer than 10 years while the large broadleaf trees had been in training almost uniformly for 20+ years. The conclusion I draw is that it takes more time and dedication to grow good broadleaf material than it does to create stunning bonsai from collected conifers. Perhaps in The Artisans Cup in 2020 we will see a broadleaf tree in the top 5 in the scoring. Will it be yours?