Wouldn’t Denny’s make a great sponsor for these curling
events?
Welcome to the 2011/12 Curling Season! Ok, I know, it started months ago, back when
most of us were still on the golf course and it runs until we tee it up
again. Usually curling for me begins
with the first televised event of the Season.
This year it was, technically, the internet coverage of the Pheasant
Classic in Brooks, but as my time was limited in watching, I’ll begin with this
year’s first Grand Slam, the GP Car and Home World Cup. Or GPC&HWC GS, for short.
Glenn Howard faced Mike McEwen in the Quarterfinals and
though not always a display of great shotmaking, it was an entertaining match.
The 7th end has Team
Howard up one without last rock. Howard
is unable to add pressure and create any 4 foot play while McEwen is able to
keep things clean and sits one back four with two shots remaining.
Howard is Yellow
Howard’s squad spends some time together in the house,
discussing alternatives to what would appear to be a hit and stick. However, this would allow McEwen to blank and
give him hammer, and a chance to score two and win in the final end. The situation has become very common when 1
up without in the next to last end: be aggressive and attempt to either steal
or force your opponent to one, at the risk of giving up two. This exact situation (open rock, back rings)
was examined in the early days of Curl With Math (http://curlwithmath.blogspot.com/2007/01/aggressive-play-in-9th-end.html).
My assessment back
then was that you need to surrender a deuce no more than 20% of the time to be
the correct call. What was interesting
in the dialogue, other than the fact it took them so long, was the comments “no
one has scored a deuce, so they probably won’t get one in the last end”. Even said with a smile by Craig, this comment
does not base any of the weight in the decision on rational thinking. The “gut” says “we haven’t scored much,
neither have they; let’s head to the next end up one”. The problem is, unless ice or rock conditions
are a factor, their chances of scoring two next end is not related to how many
deuces they have scored during the single game or even past several games. This is comparable to a baseball manager
deciding to bring in a relief pitcher against Albert Pujols simply because he’s
batting 0 for 5 against him. His chances
of belting a homer in his sixth appearance against the pitcher are no less
likely than if he were 5 for 5.
So after much discussion, Howard made what I agree was the
correct decision. In future, perhaps
this type of call won’t take as much deliberation. Contrast for example with Mike McEwen’s
decision in the next end. One down and
facing this with his final shot:
McEwen is Red
Rather than play a simple draw for one and go to the extra
end without last rock, McEwen elected to try a hit on his rock outside of the
rings, and roll his shooter into the yellow stone, to possibly score two and
win the game.
Mike correctly understood that his chances of stealing in an
extra end against Howard are not good.
Stats for Grand Slams have shown that teams tied without and one end
remaining only win 20% of the time (larger sample size shows closer to
25%). See “Statistics for Grand
Slams” at the bottom of http://curlwithmath.blogspot.com/2008/11/statistics-for-womens-curling-and-what.html
for more details.
Mike Harris commented he felt McEwen would make the shot 20%
of the time, which puts it at roughly an even decision. What Mike Harris forgot to factor in was the
analysis of still taking one, even if they miss the deuce. From Mike McEwen’s comments (writing as
tucker22) in a recent post on curlingzone.com http://dev.curlingzone.com/showthread.php?s=b918da19a44a9ff96b8c96ab365cf9fd&threadid=8894&pagenumber=3&forumid=10&tp=0
“Here's what was going through my head:
#1) VS the best 5 or so teams in the World you have only a 10-20% chance of stealing depending who you are playing. Yes, it can get as low as 10% vs the best (seen the stats).
#2) In our minds we make contact on that in off 40-50% of the time .... and we actually stick it for 2 maybe 30% of the time give or take.”
#1) VS the best 5 or so teams in the World you have only a 10-20% chance of stealing depending who you are playing. Yes, it can get as low as 10% vs the best (seen the stats).
#2) In our minds we make contact on that in off 40-50% of the time .... and we actually stick it for 2 maybe 30% of the time give or take.”
I expect the numbers showing 10% are a small sample size and
Mike is not giving his team enough credit.
I would put them closer to 20% at a steal chance, but still you are
losing 4 times for every 1 victory. What
was also interesting in the discussion was clear understanding from the team
that this was the shot, likely decided before it had come to rest. The time was spent on how to play it, not on
whether to play it. Great call,
unfortunately poor result.
Joan McCusker said they “gambled and lost”. Another example of how this phrase gets used
incorrectly. Which do you think is the
greater “gamble”, attempting to win with this shot or taking worse odds in the
extra end without last rock?
The final game between Epping and Howard was one I watched
with speed and determination, using the fast forward setting on my Tivo. Howard successfully used a corner guard in
the first end to score a deuce (despite some great shot making by John and
third Scott Bailey), but Epping bounced back with a deuce to tie (even with only
1 free guard!). This game had a more
than normal amount of corner guard play and after a miss by Epping in 4, Howard
was able to use another one and score a three to take a lead he would not
relinquish. Even if Howard only scores a
deuce, they should win at least 79% of the time. The three increased their chances to
90%. Here is a clear case where possibly
the 5 rock rule could give a team some more hope of winning (2 or 3 down with 4
to play), but how much is yet to be determined.
The 5th end was a prime example of Howard able to remove the
corner guard and Epping having to make freezes on open rocks in the house in an
effort to score two (and does), but he never really had any chance to tie with
a three. The 6 and 7th end
played out just as they will under a 5 rock rule. The downside being that the team without
hammer, when down, gains no additional advantage over today’s rules. Must say I’m more than mildly interested in
the next Slam to see how this type of end will be played. Perhaps we need more variation in the rules
of each event to create additional interest.
But please, I beg you, no mixed doubles.
I don’t know if it’s Wayne Middaugh’s influence but this Team Howard is not dressing to match their colour
tone (very “spring” and most of them are “winter”). That bright green with yellow trim does not
match anyone’s complexion. It will,
however, ensure they make it safely across the street during the late night hours;
cars will easily be able to see them from blocks away.
It won’t be all blogging for me this year. My 7 year old son actually watched the
Scotties and the Brier last year and decided it looked cool enough to try this
season. This is surprising given he
mostly gravitates to skateboarding, snowboarding, BMX and any other extreme
sport that gives his mother a nervous breakdown. My boy’s curling enlightenment led me to haul
out my dusty curling pants (is there a slicker pair of black slacks that you
can own?) and shoes and join the local club (St. Albert).
I’ve put myself on the spare list and have memorized my line of “Sorry,
I’m out of town that day” already, so as to ensure no actual rocks will be
thrown. Alas, I did agree to help out
with the Sunday Junior program and my body already hates me for it. The good news is that not just my knee is
prepared to give out, but most of the muscles, joints, tendons and other items
which make up my legs as well.
It is interesting that they call the program “Little Rocks”,
but in fact don’t use the actual little rocks; lighter stones which are thrown
by little persons, to a house that is shorter than the normal size rink. When a 4 foot kid starts at basketball, do
they play on a 10 foot hoop? Does a
tee-ball player face an 80 mph fastball in his first trip to the plate? The idea of having 7, 8 and 9 year olds
trying to slide/push/move a 40+ lb rock down the sheet is ludicrous. It’s surprising to me that most of the kids
have not become more discouraged. I hope this interest is maintained but don’t
believe we are properly setting the stage to bring these young players into the
game.
I was tempted to take my son out of the program when I saw
that they were attempting to play the adult version of the game. I started at 12 and given the physical
barriers, would suggest there is no need to start anyone before 11 unless the
smaller rocks/rules are used (except perhaps your child eats hormone enhanced
chicken and already shaves). He remains
interested for now and I will support his efforts, with the hope that his
continued failure to get a rock that weighs slightly less than he does to the
hog line, much less the button (at the other end) will not discourage him
enough that he decides to drop curling for more snowboarding.
Hi Kevin,
ReplyDeleteEnjoy your blog - always great reading. I strongly agree with you about Little Rocks; I also believe that it is unreasonable to have a "little rocks" program for kids 10-and-under when they must throw granite stones on a full sheet.
In your post, you wrote: "This is comparable to a baseball manager deciding to bring in a relief pitcher against Albert Pujols simply because he’s batting 0 for 5 against him. His chances of belting a homer in his sixth appearance against the pitcher are no less likely than if he were 5 for 5."
That would be true if the events were independent (statistically speaking) but they are not. In this case, past performance may well be indicative of future performance because these are humans, not coins or other random choices. To give a counterexample from baseball - while a hitter's batting average is their average over all at-bat's (AB), it is almost universally true that every hitter's average with runners in scoring position (RISP) is higher. There are a multitude of factors that influence that higher average: the hitter is more compelled to put the ball in play to score, the pitcher must pitch from the stretch, and often the infield is playing "in", making it easier for the hitter to get an infield hit. Consequently, a player who plays on a team with other good hitters is more likely to have a higher average than he would if he played on a mediocre team, because the events of the game are not independent and involve more than simply the hitter/pitcher combination.
I appreciate your efforts at examining my example but, please also recognize it's an example, current sabremetrics have proven that a sample size of 5 at bats of pitcher versus hitter are not indicative of future performance. Some good read's on this topic are "Mathletics" by Wayne Winston and "The Book" by Tom Tango.
ReplyDelete