Jiro Ono and René Redzepi have a Cup of Tea

Having watched the excellent Jiro Dreams of Sushi and seen René on several episodes of The Mind of a Chef, this video is both intriguing and charming.

I'm not sure I fully understand the drive that these men have to better themselves. In all honesty, while I like to think I am self-improving day by day and year to year, I'm sure that some basic level of laziness stops me from progressing as much as I am able.

Also, every time I see Jiro serve sushi it makes me want to snap book a trip to Japan.

Simulating the NHL Playoffs 2014-2015

For those who have followed me on my old blog, I have a simulator that takes some of the stats from the regular season to predict what will happen in the playoffs.

This year, I finally put a little time rewriting the simulator. Instead of using the total goals for and goals against, I used GF60 and GA60 from hockey analysis. Unlike my old blog, I think I'm going to keep updating this post with each successive round just to keep things a little more organised.

Round 1

(Posted on April 15, prior to the first round of play).

So here are your first round predictions (including chances of reaching the later rounds). I also included the final SportsClubStats (SCS) win cup % for comparison purposes.

Reach Rd 2 Reach Rd 3 Reach Final Win Cup SportsClubStats
STL 55.44% 25.19% 15.72% 7.73% 10.70%
MIN 44.56% 17.91% 10.37% 4.55% 5.00%
NSH 60.34% 36.91% 25.91% 14.98% 4.30%
CHI 39.66% 19.99% 12.04% 5.70% 6.80%
ANA 41.90% 25.62% 9.25% 3.42% 2.80%
WIN 58.10% 39.78% 17.57% 7.94% 4.20%
VAN 45.77% 14.81% 3.59% 0.92% 3.50%
CGY 54.23% 19.78% 5.55% 1.60% 4.60%
MON 61.91% 29.12% 14.80% 7.59% 6.40%
OTT 38.09% 13.29% 5.11% 2.01% 3.50%
TBL 71.27% 45.46% 27.25% 16.53% 13.10%
DET 28.73% 12.13% 4.63% 1.81% 2.50%
NYR 65.65% 44.93% 25.93% 15.63% 21.10%
PIT 34.35% 18.05% 7.62% 3.40% 1.80%
WSH 52.40% 20.13% 8.20% 3.55% 7.20%
NYI 47.60% 16.89% 6.46% 2.63% 2.60%

Some pretty interesting spots. Tampa Bay comes out as the surprising favourite to win the cup, thanks I think primarily due to its first round matchup against Detroit (This is not the Detroit team of old, at least according to their GF60/GA60 stats).

Nashville is the clear favourite in the West due to their very strong goal differential. Comparing to SCS, it's pretty obvious that whatever they use for their simulations does not use goal differential as the primary simulation mechanism. I'm pretty curious now what they do use.

Winnipeg looks like a bit of a sleeper pick from these numbers - I think largely due to their relatively soft bracket (getting by Anaheim is a bit of a challenge, but then a matchup against Vancouver or Calgary is a good spot to be in).

And the Rangers look like big favourites to come out of their division. So will we see a NSH v WIN and TBL v NYR semifinals?

Round 1 Results (posted April 30)

Reach Rd 2 Reach Rd 3 Reach Final Win Cup SportsClubStats
STL 55.44% 25.19% 15.72% 7.73% 10.70%
MIN 44.56% 17.91% 10.37% 4.55% 5.00%
NSH 60.34% 36.91% 25.91% 14.98% 4.30%
CHI 39.66% 19.99% 12.04% 5.70% 6.80%
ANA 41.90% 25.62% 9.25% 3.42% 2.80%
WIN 58.10% 39.78% 17.57% 7.94% 4.20%
VAN 45.77% 14.81% 3.59% 0.92% 3.50%
CGY 54.23% 19.78% 5.55% 1.60% 4.60%
MON 61.91% 29.12% 14.80% 7.59% 6.40%
OTT 38.09% 13.29% 5.11% 2.01% 3.50%
TBL 71.27% 45.46% 27.25% 16.53% 13.10%
DET 28.73% 12.13% 4.63% 1.81% 2.50%
NYR 65.65% 44.93% 25.93% 15.63% 21.10%
PIT 34.35% 18.05% 7.62% 3.40% 1.80%
WSH 52.40% 20.13% 8.20% 3.55% 7.20%
NYI 47.60% 16.89% 6.46% 2.63% 2.60%

If you picked the team that the simulator gave a >50% edge to, we would've gone 5 for 8, with all the misses in the Western conference. A bit of a rough start to this playoff season. Let's see what we do for round 2!

Round 2

(posted on April 30)

Reach Rd 3 Reach Final Win Cup
MIN 0.467262 0.284583 0.114486
CHI 0.532738 0.338333 0.1484
ANA 0.592412 0.243603 0.081402
CGY 0.407588 0.133481 0.034306
MON 0.409253 0.191011 0.108007
TBL 0.590747 0.329915 0.218204
NYR 0.669791 0.358398 0.236204
WSH 0.330209 0.120676 0.058991

The cup favourites are now both Eastern teams: Tampa Bay and the Rangers both enjoy huge edges in this round. The West is a bit more random with Calgary having a pretty uphill battle, and Chicago given a slight edge to win the cup.

Should be an interesting 2nd round. As a hockey fan, I'll be rooting for the Canadiens, although I think they're in tough against the Lightning.

Round 2 Results (posted May 14)

Reach Rd 3 Reach Final Win Cup
MIN 0.467262 0.284583 0.114486
CHI 0.532738 0.338333 0.1484
ANA 0.592412 0.243603 0.081402
CGY 0.407588 0.133481 0.034306
MON 0.409253 0.191011 0.108007
TBL 0.590747 0.329915 0.218204
NYR 0.669791 0.358398 0.236204
WSH 0.330209 0.120676 0.058991

This time around the simulator picks the winner as the favourite in all four series, so we improve to 9 for 12 overall so far. It almost wasn't the result though. Washington gave the Rangers a pretty tough go of it. Despite the Rangers having serious difficulty scoring goals they managed to eke out three consecutive wins to take the series back. Let's check out the next round shall we?

Round 3

(posted on May 14)

Reach Final Win Cup
CHI 0.6018 0.241522
ANA 0.3982 0.117455
TBL 0.50152 0.322258
NYR 0.49848 0.318765

The simulator thinks that an Eastern team will win the cup this year with greater than 60% shot to win the cup. Who will it be, that is the question!

I'm involved in two hockey pools this year, and my live teams are the Lightning and the Rangers (one in each pool). So I'm guaranteed a finals shot in one of my pools. Here's hoping I win one!

Round 3 Results (posted May 31)

Reach Final Win Cup
CHI 0.6018 0.241522
ANA 0.3982 0.117455
TBL 0.50152 0.322258
NYR 0.49848 0.318765

It went to the wire in both series, but the simulator "picked" the winner in both series (although the Tampa Bay was pretty close to a straight up coin flip). So we improve to a 11 for 14 prediction rate for this round.

Stanley Cup Finals

(posted on May 31)

Win Cup
CHI 0.400347
TBL 0.599653

So the simulator picks Tampa Bay as 60% favourites for this finals. I backed Tampa Bay in one of my playoff pools, so here's hoping they take it down!

Final Results (posted June 17)

Win Cup
CHI 0.400347
TBL 0.599653

So Chicago wins their third cup in six years and basically has a modern dynasty on their hands. Congrats to them, even though it cost me from winning my playoff pool. Ahh well. To my eye, the Lightning got a little unlucky in a couple of those games.

Anyway, the simulator goes 11 for 15 this year or 73.33% which isn't too bad. I've been doing this for awhile so previous records look like this. Cup winners and original simulator prediction of cup win % included for interest's sake.

2015: 11/15 (73.33%) - Chicago Blackhawks (5.7%)
2014: 7/15 (46.67%) - Los Angeles Kings (4.6%)
2013: 12/15 (80.00%) - Chicago Blackhawks (36.8%)
2012: 7/15 (46.67%) - Los Angeles Kings (2.5%)
2011: 11/15 (73.33%) - Boston Bruins (13.6%)
2010: 10/15 (66.67%) - Chicago Blackhawks (16.6%)
2009: 8/15 (53.33%) - Pittsburgh Penguins (3.6%)
2008: 14/15 (93.33%) - Detroit Red Wings (33.0%)
Total: 80/120 (66.67%)

See you next year!

Talented 15 Year Old Guitarist

This is an incredibly talented fifteen year old guitarist. Dragon Force is a metal band known for their blisteringly fast guitar riffs and solos. This particular song was featured as the most difficult song on Guitar Hero 3.

Just incredible. She has a pretty impressive collection of other songs on her YouTube channel, if you want to see more.

My Favourite Things of 2014

2014 was a fun year, and I thought I'd summarise it by reviewing some of my favourite products of the year. These products fall into a few categories.

Apps

mzl.pdumvzwl.png

Overcast - by Marco Arment

As you'll see in the podcast section, this really was the year of the podcast for me. I started picking up podcasts before Marco Arment finished Overcast. However, when it came out with its groundbreaking features of smart speed and voice boost, Overcast quickly become my podcast listening app of choice. Smart speed really is the killer feature of this app: it strips out a portion of gaps in the audio. Over the course of a couple hour podcast, this feature can save a very significant amount of time. At the time of writing, Overcast has saved me 61 hours of dead air. I may have gone a bit crazy listening to podcasts ...

Monument Valley - by Ustwo Games

This combination game and art piece launched on iOS on the 3rd of April this year. It's fairly short, with none of the ten levels in the original game being particularly time consuming or challenging. This continues through the nine additional levels added later in the year. But each level is a masterpiece. I spent my playthroughs of the game oohing and aahing as each level progressed.

One of the big reasons I loved this game as much as I did was the obvious inspiration from the art of MC Escher who happens to be my longstanding favourite artist (ever since I read Gödel, Escher, Bach). This game is definitely worth the few dollars it costs for both the main game and its expansions. I fully expect to return to this masterpiece of a game over and over again.

Threes - by Asher Vollmer, Greg Wohwend, and Jimmy Hinson

This game was released on iOS way back in February this year, and is still a huge goto game for me. As a mobile game, it is a great way to kill a few minutes here or there. The game has a simple mechanic to push squares around the board, and quickly became very addictive to me. Both Diane and I have spent an inordinate amount of time playing this game, and it might very well continue to be the game I keep playing through next year.

My high score so far is 79,167 which I'm fairly happy with. But it sure would be nice to get that 3072 and then that 6144 tile ...

Board Games

Not all these games may have come out in 2014, but instead have some association with the year for me.

pic1904079_md.jpg

Splendor

This was a game recommended by my favourite board game review site Shut up and Sit Down. This game has both a very simple mechanic as well as being subtly deep. The game can be taught to anyone, and is a fascinating game to play and strategise over. Just about everyone who I've played a game of this with has enjoyed it so far, so it has also played very well with my group of friends.

K2

Another recommendation for Shut up and Sit Down, this game has you competing to climb K2 - the second highest mountain in the world. The game has you managing your acclimatisation to the elevation and does a really good job of making you feel like this is a tough thing that you're doing! The mechanics are nice and easy to understand so you can jump past those quickly to playing the game with your friends.

Wok Star

This is not a new game - in fact I played it for the first time a few years ago on a visit back to Edmonton. But this year my Kickstarter copy of the game finally shipped. This is a great cooperative board game where the players are attempting to assemble dishes in a Chinese kitchen in real time. What follows is a chaos filled period that is pretty tough to describe.

Hanabi

This is a game I heard about before but didn't actually play until this year. It is a cooperative game where players do not know the cards that they hold but instead can see the cards everyone else has. The game has some mechanics to let players indicate facts about each other's hands. The game is a fun cooperative puzzle to play.

Podcasts

This was the year of podcasts for me. Prior to this year, I only really listened to the 2+2 pokercast to keep in touch with the changes in the poker industry. This year though, I've listened to hundreds of podcast episodes from a couple dozen different shows. Listening to podcasts has nearly completely replaced listening to music for me. So with so listening to podcasts, which shows are the highlights of the year for me?

Accidental Tech Podcast (ATP) - Casey Liss, John Siracusa, and Marco Arment

ATP is a very enjoyable listen on almost any episode. The three hosts put together a weekly show that entertains while they analyse tech industry news and trends and discuss topics that are of interest in the tech. Beware that the show has a fairly Apple-centric bias since all three are Apple converts - so if that is likely to annoy you then you may want to look elsewhere. But they often to not talk about Apple as well.

Analog(ue) - Casey Liss and Myke Hurley

Myke Hurley is a prolific podcaster that just started his own podcast network called Relay.fm. Of the new crop of shows, Analogue surprised me the most. On each episode, Myke and Casey explore topics that are often not talked about. They explore life changing decisions such as Myke leaving his job to go self-employed, as well how it feels to be relatively famous (for podcasters). They put themselves out there, which cannot be an easy thing to do, and as a result you get a pretty interesting listen every week.

Designer Notes - Soren Johnson

There are only two episodes of this podcast out so far, but both have been fascinating. Soren Johnson is a game designer who was lead designer on Civ IV, which I probably sunk too many hours into. His first two podcasts involve chatting with Rob Pardo who was until recently involved with game design at Blizzard: working on StarCraft on through the World of Warcraft and everything in between. The podcasts are a fascinating listen on the world of video game game design.

Hello Internet - Brady Haran and CGP Grey

One of my absolute favourite YouTube creators is CGP Grey, so as soon as I saw him launch a podcast I jumped on immediately. This show is entertaining every single time and cover a wide variety of topics. When this show is released it bubbles quickly to the top of my queue to listen to.

Isometric - Brianna Wu, Georgia Dow, Maddy Myers, and Steve Lubitz

This is a latecomer to my podcast listening in 2014. I listened to the episode where John Siracusa joined them to talk about Nintendo, and then started listening to the show as it came out ... and then decided to go back and listen to all the old episodes which I did over the course of a couple of weeks. This is a video game podcast that is really entertaining - the hosts have a great sense of humour. At the same time, they cover a lot of serious topics like the lack of diversity (both in gender and race) in video games. Some of the episodes from around the time that gamergate was exploding are a bit hard to listen to since a couple of the hosts were among those getting death threats. They did a very good job explaining what was going on and how it was personally affecting them.

Serial - Sarah Koenig

Unless you've been hiding under a rock, you've probably heard of Serial the podcast. It was a very interesting listen - particularly at the start. Sarah Koenig explores a particular real-life murder case where a high school student is convicted of murdering his ex-girlfriend. What's crazy is that the prosecution case seems extremely thin, and it is a bit scary that he was jailed on such a light set of evidence. However, throughout the show you are left wondering whether the guy actually committed the crime or not. The only thing that I didn't really like about the show was that I felt a bit bad that there are so many people listening to the fates of real people - and it sort of feels a bit like none of our business. However, if Adnan is actually innocent and gets out of his life sentence over this then I suppose that's worth a fair amount.

Tools and Gadgets

These are some of the Kickstarter projects that came in for me recently that I want to share.

GIR spatulas

These fantastic silicone spatulas are extremely useful in the kitchen. Originally a Kickstarter project I backed, I loved the ones I got so much that I ordered some more and then backed their silicone lids Kickstarter as well. For anyone who likes to spend any time in the kitchen, these are great - and would make a great gift as well!

FUZ Everdocks

There are two docks that I've backed by this company. One is a dock to plugin your smartphone that sticks to your table with a microsuction grip. The dock accepts a wide variety of charging cables making it suitable for most smartphones. It is a lovely piece of design and I really like my dock. This year I received my Everdock Go Dash, which is a car dock for smartphones. I've only been using it for a couple weeks, but it is already one of my favourite products from Kickstarter.

Orbitkey

I like keeping my pockets organised, and I've long hated the ordinary keychain for being annoying in my pocket. This Kickstarter has done very well for me - my keys are always in the same configuration and no longer painfully dig into my thigh.

UofA CPRG Solves Heads-Up Fixed Limit Hold'em

The research group I was a part of around six years ago made an announcement yesterday that they had effectively solved two player limit Texas Hold'em. There is already a ridiculous amount of press to read on the subject, including Scientific American, the BBC, the Guardian, IEEE Spectrum, The Economist, The Verge, the Washington Post, and many many more.

To give you a rough idea of how impressive this is, the strategy takes 11 terabytes of disk to store, and required 900 CPU years of computing power. In order to make the game even remotely solvable and the solution storable, several breakthroughs needed to be made over the course of several years of research. The Computer Poker Research Group (CPRG) has been making progress towards this goal for a long time. The final breakthroughs include an amazing new compression algorithm for this purpose that greatly decreased the storage requirements of the solution, and an algorithmic improvement to the solver that greatly decreased the number of iterations necessary to converge on a solution.

It's important to note a couple of things:

  1. This solution is actually an approximation of the game theory optimal solution. It is actually exploitable for less than 1 milliblind/100 hands, or in poker terminology, less than 0.05 big bets per hundred hands. This is a very small amount, and it would be extremely difficult to tell this strategy apart from its best response by playing the two strategies against each other. But it is worth noting.
  2. This is only a solution for heads-up (two player) fixed limit Texas Hold'em. For the time being, other poker variants have not been solved. Other variants, such as ones where there are three or more players or where the betting is pot-limit or no-limit have not been solved. There are many challenges to overcome before these games are solved.

I plan on writing more extensively on the subject in the near future. In the meantime, I'd like to congratulate Mike Bowling, Neil Burch, Mike Johanson, and Oskari Tammelin on an incredible achievement.