Mission accomplished – Cambridge [2011-03-26]

Here are the recaps from this weekend’s Cambridge tournament. The official results can be found here.

Game #1 vs. Sharmaine Farini

Game started slowly and Sharmaine commented that I was “getting all the tiles”. After that, she switched to a very defensive style and closed off bingo lanes. She wasn’t helping herself by only playing off one or two tiles at a time.

FINAL SCORE: 420-332 (WIN)
BINGOS: Me – ArCSINE; Sharmaine – DENTIST
RECORD: 1-0 +88

Game #2 vs. Gene Rawlins

For some reason, I’ve had trouble beating Gene in the past. I decided to forget the past and just focus on making good plays. Gene had more bingos than me, but I focused on maximizing my tiles. The sequence of plays had me scoring 29, 41, 30, 18, 44, 32, 33 in consecutive turns while Gene had several <10 point plays between his bingos.

FINAL SCORE: 401-360 (WIN)
BINGOS: Me – INTERNS; Gene – LUNIEST, URInATE
RECORD: 2-0 +129  

Game #3 vs. Abdul Majid Khan

I’ve been thinking about this game since yesterday and I’m not sure if I could have done anything. I struggled with tiles which didn’t work well together while Abdul was making the most of having some good tiles. His sequence of EnCASED (68), VOX (42), JERKY (57), and TEAM (38) pretty much ended the game as he then started closing off scoring lanes systematically. Is there anything that can be done to change things-up when getting bad tiles?

FINAL SCORE: 273-464 (LOSS – ugh)
BINGOS: Me – REBOARDS; Abdul – EnCASED
RECORD: 2-1 -62

Game #4 vs. Lucie McPhail

A close game that came down to the final few plays. I will post a more detailed look at this one as I’m interested in a few board positions. My studying paid off as I played a Z7 – zOUAVES. Unfortunately, I missed a phoney 3 – NAS* late in the game that could have helped my spread.

FINAL SCORE: 395-363 (WIN)
BINGOS: Me – zOUAVES; Lucie – (T)HRILLED, RELATES
RECORD: 3-1 -30

Game #5 vs. Mad Palazzo

I’ve had trouble beating Mad in the past, but I kept a positive attitude and played confidently. The board got closed and I didn’t open things up as I had a decent 60 point lead. However, I started fishing to try and play a knock-out bingo. That strategy didn’t work as I was scoring ~10 points per turn with single tiles and Mad was making small plays that scored 15-20 points. When I did get great bingo tiles, there was no place to make a play. A bit of tile luck helped me out as I got the final blank and Mad had IRY as her last tiles. She challenged WAp(I)TI and that sealed the win.

FINAL SCORE: 367-323 (WIN)
BINGOS: Me – AN(I)SOLES; Mad – (D)OOrMATS (a great find)
RECORD: 4-1 +14

Game #6 vs. Lynda Wise

Lynda started with a bingo but then struggled to make scoring plays. I had a 100 point lead when Lynda tried a bingo: (T)YrANIES*. After a pause to write out the word, I challenged it off. I kept scoring to keep up the pressure on Lynda as I got stuck with the Q and had to eat it. Lynda also tried (STAINED)S* which was also challenged off.

FINAL SCORE: 451-408 (WIN)
BINGOS: Me – (T)ENTORIA, BELONGs, SOURING; Lynda – DORMANT, STaINED
RECORD: 5-1 +57

Game #7 vs. Emily James

Apparently we have played before. Emily mentioned that I “destroyed” her at the 2008 NSC in Orlando. I looked-up the result on Cross-Tables and it was a close game (389-375) so I’m not sure what she’s talking about. However, if that gives me a psychological edge, I’ll take it!

After playing JADE (52) and NEROLIS (67), Emily muttered something like “not again!”. To her credit, she didn’t give up and kept playing even though I I seemed to be getting the better share of the tiles. I thought I had finished the game off with BUgGI(E)ST for 72 to go ahead 376-222 but Emily took advantage of the bingo lane I created. She played QUOIN for 82 to stay within striking distance.

FINAL SCORE: 454-356
BINGOS: Me – NEROLIS, ENTICES, BUgGI(E)ST; Emily – none
RECORD: 6-1 +155

Some statistics from my seven games:

Me

Them

Average Score

394.4

372.3

Average Turns

15.1

15.0

Bingos

11

9

50+ point plays (non bingo)

1

1

40 point plays

5

3

30 point plays

17

9

Exchanges

3

1

Blanks

8

6

Challenges Won

3

0

Phonies Not Challenged

0

1

The statistics tell me that I’m not having many non-scoring turns (i.e., exchanges and losing challenges). What I found interesting was that the number of bingos and blanks was relatively even between me and my opponents. The number of 30 and 40 points plays seemed to be the difference in games. Actually, there’s something oddly satisfying about making a play and saying a number in either the 40s or 50s. I think it’s more frustrating to an opponent than seeing a bingo.

All in all, I was very pleased with the tournament. My goal was to play well but not have my rating go above 1300. Mission accomplished! Looks like I’m in good position for the Dallas in August.

Here’s a shout-out for some of the other Mississaugans who did well:

  • Joy Brown: 1st place in Div. 5. Congratulations on the win!
  • John David: 1st place in Div. 4. Congratulations on the win!
  • Sophia Ozorio: An excellent 5-2 record for 3rd place in Div 4.
  • Steve Ozorio: An excellent 5-2 record for 3rd place in Div. 2.

Also, congrats to Heather McCall for winning Div 2! Vera also had a great tournament playing with the big boys in Div 1 (5-2 for fourth place). Oh before I forget, congrats to Steven Karp for a great 5-2 record (5th place, Div 2) and also for heading off to Cambridge for more schooling.

This is pretty much it for me in terms of Scrabble for a while as I’m not planning on playing in any tournaments before Dallas. Also, I’ll be traveling to California for work so no more club play, potentially for several months. Instead of playing, I plan to study and continue building the vocabulary.

NSC By the Numbers – 1st & 2nd Place

A few days ago, I examined attendance and participation at the National Scrabble® Championship. Today, I’m going to try and answer the question: how many games does it take to win a division?

Like last time, I’ve used data taken from Cross-Tables and done some very simple analysis. Here is the data organized into a table.

2002

2004

2005

2006

2008

2009

2010

GAMES

31

30

28

28

28

31

31

1st PLACE

2002

2004

2005

2006

2008

2009

2010

Division 1

25

26

21

21

22

25

25

Division 2

24

23

22

22

24

24

23

Division 3

23

22

21

20

21

23

24

Division 4

25

27

22

22

21.5

26

25.5

Division 5

25

25

22

22.5

22

22.5

24

Division 6

25

23

22

22

23.5

Division 7

28.0

Total

24.5

24.9

21.7

21.6

22.3

24.1

24.3

Win %

79.0%

82.9%

77.4%

77.1%

79.8%

77.7%

78.4%

Avg. Win %

78.9%

2nd PLACE

2002

2004

2005

2006

2008

2009

2010

Division 1

20

23

21

19

20

25

21

Division 2

22

22

20

20

21

21.5

22

Division 3

22

21

20

20

19.5

21

21

Division 4

23

22

20

19.5

21

22.5

21.5

Division 5

23

22

20

20.5

20

22

24

Division 6

22

22

20

20.5

20

Division 7

24.0

Total

22.0

22.3

20.2

19.9

20.3

22.4

21.9

Win %

71.0%

74.3%

72.0%

71.1%

72.3%

72.3%

70.6%

Avg. Win %

71.9%

Some notes about the data:

  • The NSC was not held in 2003 and 2007.
  • “–” indicates no data
  • Ties were included as 0.5 wins
  • “Cumulative scores/differences” were not considered in this analysis
  • A ‘bye’ was included as a “win” as recorded during tournaments. This only came into play in 2010 Division 5 where the first place winner had a bye.
  • Because the number of games varied from year to year, I couldn’t do a simple mean/average across each division and instead focused on win %.
  • Also, based on visual inspection of the data, I made an assumption that the number of games required to win a division was consistent across all of the divisions. The rationale for this line of thinking is that, in theory, the division breakdown means that players will be playing against relatively equal skill for each division.
  • Did not consider “gibsonization” whereby a player has secured the win before all of the games have been completed.

After looking at the data, I was surprised at how consistent the data was across divisions and years. Of course, there are some exceptions (26, 27, and 28 wins in 2004). Across the years, the main conclusion I can draw from the data is that in order to win your division, you have to win close to 80% of your games. The second place finisher seemed to perform very well throughout the tournaments, having a win % of 72%.

So, what does this mean? For a 31 game tournament, first place means having to win roughly 24.5 games. To have that many wins is quite amazing in my opinion. Not only do you have to have skill but you also have to have a bit of luck. You also have to perform very well to get second place – 22.5 wins over 31 games is also quite a feat.

Having looked at this data, I appreciate the skill of the winners that much more.

NSC By The Numbers – Attendance

This year’s National Scrabble® Championship (NSC) in Dallas is drawing to a close and it looks like Nigel Richards has clinched yet another title. That’s quite a feat! Unfortunately, I wasn’t able to attend the NSC this year but I hope to play in next year’s event.

Anyway, I’ve been following the progress via the live coverage and some of the statistics have just fascinated me. What first jumped out at me was the attendance. Does anyone else think that fewer people are participating in tournaments lately? I’ve noticed that the “big” tournaments seem to have fewer players involved. The reason I’ve started noticing is because I enjoy the energy at the larger tournaments.

So here’s some data available from Cross-Tables for the past few years of the NSC [note: data tables will be presented at the end of this post]. From my understanding, the NSC was not held in 2003 and 2007. In Figure 1, we can see the total number of players at each NSC event.

Figure 1: Attendance at the National Scrabble Championship

The first thing that I notice is that attendance has dropped significantly since the peak of 797 in 2004 to this year’s 408. That’s nearly a 50% decrease in participation for arguably the most important Scrabble event in North America!

When we look at the breakdown by division, we see something interesting (Figure 2).

Figure 2: Attendance by Division at National Scrabble Championships

Making comparisons between years is tough because the cut-offs for divisions has changed throughout the year. For example, in 2004, there were seven divisions compared to only five for the past two years. Also, I’ve been told that the policy on “playing-up” has changed over the years. While every division has decreased in numbers (due to the overall decrease in participation), the lower divisions seem to be affected the most. Forgetting about the number of divisions, the lowest rated players don’t seem to be coming to the main event which is somewhat troubling.

So, why is this happening? I’m sure there’s no one factor, but here are some possible explanations that come to mind and that I’ve heard:

#1. It’s too expensive
With the cost of travel, lodging, and entry fees, the NSC might be too expensive for many players to attend every year.

#2. Players don’t like the venue/host cities
I’ve heard from a couple of players that *where* the NSC is played is a factor in determining if they will/will not attend. If the host city is a draw for the player (or their families), then they are more likely to attend. If the host city isn’t much of a draw for them, then they don’t go.

#3. The economy sucks
Let’s be honest. The economy has taken a toll on people over the past few years. Maybe we’re just seeing the effect now with people having less financial resources available for discretionary spending.

#4. Hasbro is no longer the official sponsor
Not sure if this is much of a factor, but Hasbro is no longer the official sponsor. Maybe this is a factor for some people.

#5. There’s less prize money
Related to #4, there is less prize money than in the past. The Division 1 champion previously won up to $25,000 and now the top prize is down to $10,000. The other divisions have seen decreases in the prizes as well.

#6. Too many local tournaments
From what some of the veteran players have said, there were far fewer tournaments in the past so people tended to play in each of them. Today, one can find a local tournament almost every month (if not more frequently). This frequency (which is great for choice) might be leading to “tournament fatigue”.

#7. People don’t like NASPA
Related to #4 with Hasbro pulling out of the tournament scene (non-school), the North American Scrabble Player’s Association (NASPA) is running the tournament scene. Maybe NASPA hasn’t been doing as good a job in promoting the NSC and growing the game over the past two years given the challenges of running and organizing the tournament scene.

Final Thoughts

I don’t think there’s any one reason why people are participating in the NSC less than in the past. Declining participation is probably due to a number of different factors. I don’t know if the overall number of players participating in tournaments is increasing, decreasing, or staying stable – which might help to explain some of the attendance. Were the last two years of NSC attendance a blip or is this the new normal?

In the next few days, I plan on doing an analysis of what it takes to win a division at the NSC.

Here’s the data from Cross-Tables.

2002 2004 2005 2006 2008 2009* 2010*
GAMES 31 30 28 28 28 31 31
ATTENDANCE 2002 2004 2005 2006 2008 2009 2010 Average
Division 1 132 173 87 76 103 123 116 116
Division 2 103 139 108 102 134 90 88 109
Division 3 136 169 135 129 133 128 98 133
Division 4 126 150 124 116 116 93 64 113
Division 5 111 90 94 88 96 52 42 82
Division 6 88 76 134 114 80 98
Division 7 40 40
# of Players 696 797 682 625 662 486 408

A look at ratings #2 – the ELO system

In a previous post, I took a look at some of the statistical correlations between a player’s scoring averages and their Scrabble® rating.  Over this past year, I’ve seen my own personal rating jump by over 400 points (in tournament play) and about 200 points at club play.  Seems like everyone likes to talk about ratings – either theirs going up or down.

In this post, I’m taking a closer look at the rating system.  For those of you who don’t know, here in North America, the National Scrabble® Association (NSA) uses the ELO system originally developed and still used for chess.  I won’t go into too much detail, but this system is based on estimating a player’s "true skill" by using statistical measures of interpreting wins & losses.

The ELO rating system
Basically, each player has a rating.  This rating is then compared with your opponent’s rating and an "expected" win % is determined based on how far apart you and your opponent’s ratings are.  After each game is played, you calculate the difference between the expected % of wins and our performance.  If two players are equal in skill, then in theory, they should each win 50% of the time against one another.  A higher rating would thus suggest higher skill and higher likelihood of winning any particular game against lower rated players.  The same works when playing against higher rated players.  The theory is that over time, your true skill should be reflected in your rating because your "good and bad" playing should even out over time and your true rating should emerge.

For a more detailed explanation of the ELO system, please view the Wikipedia entry titled "Elo rating system".

Complaints about the rating system(s) in Scrabble
In the recent issue of Scrabble News (#218 I believe), there was a note about the formation of a ratings committee.  This group is going to suggest an alternative ratings formula and come-up with something for use by 2009.  I’m not sure why people feel a need to change the current system, but I’ve heard the following complaints:

  • Ratings have been decreasing/"deflating" (read below for some more info about this)
  • Point "spreads" should be acknowledged – for example a player winning by 500 points should be recognized
  • Similar to the point above, "all wins aren’t equal"
  • It’s too easier to "lose" rating points but harder to gain them

Personally, I’m not sure what all the complaining and clamoring is all about.  No system is perfect.  In science, it’s what we call "measurement error" in that every measurement we take has some sort of error involved.  For things like measuring the height of something, we can be pretty good, but with things like measuring someone’s intelligence, it can be hit or miss.

Ratings deflation in Scrabble®
In the case of people complaining about ratings "going down", the Wikipedia entry has a very interesting example that describes this phenomenon of ratings decreasing even though a player’s skill level remains the same (it’s about halfway through the section titled "ratings inflation and deflation").  Because of the way the system is set-up, if only a few players improve their skill while everyone else stays the same, what happens is that the player who got better gets a big jump in ratings while the others who stayed the same decrease in ratings – thus the "deflation".

I believe this is what is happening in Scrabble today.  Players who used to be in the 1500-1600 range have had their ratings drop a good 200-300 points in the past few years.  What I’ve seen as a new player is that more players are getting better and at a quicker rate.  We can see this in some of the younger players getting involved, but mostly because of the better study tools available.  Computer simulation software allows players to analyze games and become better.  We also see this in the decreasing number of (active) players in the 1900s and 2000s.  What we don’t see, however, is a change in the relative rankings of players.  The top players are still at the top in roughly the same order as they were before.  It’s those players who haven’t improved much over the past few years who have seen their ratings plummet.  They aren’t any worse than they were before.  It’s just that everyone else is getting better and so gaining and/or maintaining a rating is much more difficult.

In my example, based on my play at my local club, I’ve felt that I was about a 1150-1250 player and yet my tournament/NSA rating was about 800 for over a year.  Once I figured out a few things about playing in tournaments, particularly about playing against lower rated players, I’ve seen my rating jump 400+ points to about 1205 (estimated).  I’ve jumped up so much because my true level of play is higher than 800.  Now that I’m at the 1200 level, I don’t expect to see much gains in my rating until I start improving again.  Those people who I beat would see their rating go down, not because they played poorly, but because they played someone who was rated lower than them.  Similarly, I would expect it very difficult to keep my rating if I start to play against opponents who are getting better.  We’re all on this treadmill of needing to improve or else see our ratings go down.

What people also complain about is the fact that unlike in chess, Scrabble® involves quite a bit of randomness and luck.  Sometimes you get some crappy tiles and it’s almost impossible to win.  Unfortunately, there’s really not much to do.  In the card game Bridge, the solution was to create "duplicate bridge", where every team plays the exact same hands/cards.  The theory is that if everyone plays the same cards/hand then those with better skill should do better.  It’s like having everyone take a standardized test and getting a score.  Unfortunately, I don’t think this is likely to happen in Scrabble.

Final thoughts
So after a brief look at the ELO system, I’m not sure if any new system that may (or may not be introduced) will fair any better.  Personally, I think the secret is not to worry about the rating and focus on just becoming a better player.  The ratings will take care of themselves.

Powered by Qumana

What is a bingo worth?

As I’ve written about the frequency of bingos, I’ve received some interesting feedback.  Most of the people who have spoken to me have had similar thoughts – they expected to bingo more.

I’ve also done some analysis about the correlation between scoring and ratings, concluding that a better player scores more points.  Sure that doesn’t sound too earth-shattering, but I have some data to support my hypothesis.  That’s probably why I’ve been trying to learn more words and also trying to bingo more frequently during games.  Bingos are the fastest way to score points.

Okay, so what is a bingo worth?  When I play my games, I often do mental "guesstimates" of how many bingos ahead/behind I am.  Here are the rough values that I use for 7 letter bingos:

Regular bingo:  60 points
Bingo + DWS:  70 points
Bingo + TWS:  80 points
"double-double" bingo:  90 points
"triple-triple" bingo:  140 points

Just to clarify, DWS means "double word score" and TWS means "triple word score".  A "double-double" is a play when two DWS squares are used and the "triple-triple" is a play when two TWS squares are used.

I mentioned these values to Craig Rowland and he thinks my figures are a bit on the high side, especially for the regular and DWS bingos.  These values aren’t meant to be perfect, but I use the to give me a sense of what needs to be done to catch-up to an opponent have an opponent catch-me.  Of course, if you are lucky enough to have some premium tiles included with your bingo, then the point value can jump up considerably.  But overall, I think the values are reasonable.

Of course, there is also the psychological value of a bingo:  it can demoralise an opponent and give you a boost, not to mention having greater tile turnover (and thus better chances at getting the better tiles).  Having been the recipient of a few bingo-bangos (that’s two bingos back to back), I know that it’s pretty frustrating and even demoralizing.  I can only imagine what it must feel like to get three bingos played successsively played against me (i.e., a bingo-bango-bong).  On the other hand, playing successive bingos is such a cool feeling and you can see your opponent’s body language just slump.  After playing a bingo, I always feel better – as if I achieved something.  It’s like making a clutch put in golf.

What do you think?  Are these values too high?  Too low?  Personally, I think the value of a bingo is much more than points.

Powered by Qumana

How often do you bingo in Scrabble®?

How often do you bingo during a typical game of Scrabble®?  Lately, I’ve been thinking about this question.  Maybe it’s because I’ve started studying words and am trying to a) justify the time spent studying and b) find a more efficient and effective way of studying.  As I previously wrote, we know that dedicated, systematic, and structured preparation (i.e., study) can lead to what we call "expertise".  But what and how should one study?

A few years back, Craig Rowland (director of the Mississauga Scrabble Club) held an afternoon of "Scrabble School" during which he shared a recommended study strategy.  I’ve been studying so that I can try to bingo more frequently because it’s the fastest way to score points – if you can get lucky, you can score huge points.

Anyway, I asked Craig how often he, and other experts, bingo in a typical game.  I expected him to say two or more, but he surprised me.  He said that on average, he bingos less than two times a game (something like 5 bingos in 3 games or 12 in 7 games), and would expect the same for other experts.  Maybe I only recall the anomalous games when a player has 4 or 5 bingos, but I honestly expected experts to bingo more than twice a game.  Granted, the examples he gave suggest something very close to 2 bingos a game, but not exactly.  Unfortunately, I didn’t have any data to back-up my assumptions.

I did a bit of digging and have compiled some statistics for the 32 games I’ve played in 2008.  Here is the data:

  • My average bingos per game:  1.59
  • My opponents bingos per game:  1.22
  • Combined bingos per game:  2.81

That’s a bit less than even I expected.  I plan on collecting this data for all of my games played and will see what happens.  I’ve been asking around at the club, and a few experts said they expected two per game, but then changed that to "under two" being a more realistic value after thinking about it.  I’d love to learn how others do – feel free to share your statistics/numbers below.  I’ll post updates on my figures later in the year to see if my studying increases the number of bingos I play.

Powered by Qumana

2007: A year in review

Since I had my wisdom teeth removed yesterday (Wed. Dec 12th) and don’t think I’ll be playing any games this week, I thought I’d review some of my accomplishments for 2007.  Here’s a quick review of this past year.

Achievements/Milestones/Highlights:

Based on the stats at Cross-tables, I’m currently tied for 28th for most ratings gains over the past 12 months (+365).  I hope I can stay on this list in 2008.

Looking back at my tournament results, I still have much to improve.  I definitely need to become more consistent – I can’t seem to play well in  consecutive tournaments.  I’ve been doing some more analysis of my games and trying to find strengths and weaknesses of my game.  Because of my research/school work, I haven’t been able to study as much as I had hoped.  I’m hoping that I can do more study and analysis in 2008.  Using tools like Quackle and Zyzzyva have really helped me improve my game.

Looking back, I found these two posts to be very helpful.

  • A closer look at Scrabble ratings – A simple analysis based on data from the club.  Basically, if you want to get better, you need to be able to score more – higher rated players tend to score more.
  • How to become a champion – Sometimes I forget the tips I got from Word Freak.  It’s always good to review these.  I’m still working on improving my control and discipline during games.