Monday, June 28, 2010

Rox Talk - Week 12

The Week That Was
A 3-3 week against the Red Sox and Angels. The Rox currently stand at 39-36. Six games back from division leading San Diego in fourth place. Currently 23-14 at home and 16-22 on the road, the Rox have outscored its opponents 341-306 (expected wins is 42 versus historical wins at 35). On pace for 84 wins with 737 runs scored and 661 runs against. Playoff indicator (Runs scored/Runs Against ratio) is 1.11 (>1.16 indicates high playoff potential).

At the almost half way point I think we can safely say that if the Rox don't sweep the series in San Diego this upcoming week their season is done. Currently six games behind the Padres, any hope to getting back into the race would be heavily dependent on the Rox taking three games and getting to within three. Three losses and we would be nine out. With a series then at home against the Giants followed by St Louis and finishing up with San Diego before the All Star break these 13 games are the season. The real sad part in all of this is that pitching as been masterful (go figure the staff has allowed 153 runs both at home and away). Perhaps when this season is over we will look back and consider this to be the greatest set of hurlers in Rox history. Too bad the offense decided to take a year off (see below). Hate to dismantle the team but perhaps it is time to let some members of this team go and try to get some future power in return.

Projected Runs Score
Woody Paige was on the mark last week in his article about the total lack of offense this team has produced. The team has been highly deficient in the power department. With a team full of potential big bats the lack of production is just too great to overlook anymore. With the team, management, and owners still believing that we just haven't gotten hot yet mentality I see this season becoming very much like 2008. Good teams can't wait to go on long winning streaks. They make up for this by taking 2 of 3 or 3 of 4 out of every series. These Rox have had brilliant moments but they fail to capitalize on bad teams. Their pitching continues to keep them afloat but having the 3rd most strikeouts in NL behind Arizona and Florida would suggest that "hot" is a relative term and that we might get "warm" which could mean a Wild Card run but most likely a .500 finish.

As mentioned above the Rox are on pace to score 737 runs. This value is simply the current runs scored (341) times number of expected games played in a season (162) divided by the number of games played (75). The question then is how closely will the Rox come to this number at the end of the year? If you take the previous 17 years (take out '94 and '95 seasons since they did not play a full schedule) and calculate the number of projected runs scored through the season by game number and then subtract this by the actual run score value you get a number of runs difference. So for instance in 2009, the Rox after Game 10 were on pace to score 842 runs. In actuality they scored 805 in 2009 so the difference was minus 37 runs after ten games. In the other years the values at Game 10 ranged from more than 390 above actual to 261 below actual valve. If you average these differences for all the years at Game 10, you end up with an average of 19 runs above average. This average is the blue line below.

The blue indicates that early in the season, though about 36 games the average difference between actual and projected is negative or in other words the runs score pace is higher than what becomes average. From that point it fluctuates but settles in. The error bar lines on the graph is the standard deviation from the blue point valve. Obviously early in the season when you score 10 runs or 0 runs it can cause your projection to fluctuate to the extremes but then as the season settles in your difference becomes less. Either way the data would suggest that Rox 737 projected runs is here to stay. Knowing that our run score environment is in this range it would mean if we want to be a playoff team our allowed runs need to be in the 635 range. So if our offense doesn't start to pick up our pitching as to get stingier. Good luck with that!

This team has shown instances of being the juggernaut it was suppose to be but with injuries and the continuous shuffling of the batting order it would suggest that realistically this team isn't going to wake from its offensive slumber. Not to be negative but it would seem a mini rebuilding scenario is a likely path forward for 2010. Look at the D-backs, a similar team with a potent future that never lived up to expectations. Do the Rox really want to be in last place next year? Couple of arm injuries and that is where we will be!

And Finally
Check out my article at Fangraphs.

Monday, June 21, 2010

Rox Talk - Week 11

The Week That Was
A 3-3 week against the Twins and Brewers. The Rox currently stand at 36-33. Four games back from division leading San Diego in fourth place. Currently 21-13 at home and 15-20 on the road, the Rox have outscored its opponents 311-269 (expected wins is 39 versus historical wins at 32). On pace for 85 wins with 730 runs scored and 632 runs against. Playoff indicator (Runs scored/Runs Against ratio) is 1.16 (>1.16 indicates high playoff potential).

The Rox win some, lose some. They just seem to be oscillating between a couple game. Last year at this time the Rox couldn't lose. Two years ago they couldn't win and so this year we seem to be in between. Looking at the schedule I suddenly see a really long stretch ahead for the Rox unless they suddenly find their offense. Three with Boston, three with Angels, three with Padres, four with the Giants, three with St Louis, and finally three with the Padres before the All Star break - that is just an insane schedule. If Rox make it to the All Star break with any sort of winning record then they will definitely be in the mix in September.

Ubaldo Jimenez - Once in a Lifetime Season?

If you had told this Colorado Rockies fan ten years ago that our team would have a pitcher who could possibly start the All Star game and then possibly win a Cy Young, then this fan would either say you were crazy or believed the Rockies had moved to another city. I don’t pretend that our team is the center of the baseball world rather I know the Colorado Rockies are stuck in no man’s land. We are neither East Coast nor West Coast. Our team is rarely seen and our players simply don’t get the respect they deserve due to the Nintendo Ball that was played here in the 90s. Why do I bother with such an introduction?

Well I think this explains the case of Ubaldo Jimenez. On April 17, Jimenez became the first player in franchise history to throw a no-hitter. Jimenez’s story was a feel good moment for the Colorado Rockies. Jimenez is a nice kid, with a fast ball like no other, pitching for a team where pitchers go to die. The media gave him his due and moved on to Oakland’s Dallas Braden. But this was only the beginning and Jimenez has since then rattled off ten more wins. At 13-1, he has done something only two other pitchers can claim to have done in MLB history. Sometimes this doesn’t fit well with the baseball media establishment. This was supposed to be the year Roy Halladay was going to sweep into the NL and blow batter’s away. Baseball although sometimes works in mysterious ways and Colorado should definitely not have a pitcher that can be so adverse to runs (only 13 so far). So what tends to happen? Articles start to sprout up touting the statistical mumbo jumbo that Jimenez is simply lucky, that it is all a smokescreen, and that eventually the stats will catch up and he will be revealed as an imposter. That is the funny thing about stats, when the outlier shows up, the men behind the numbers rationalize away the beauty of baseball, and either discount the player or the situation. The all telling models have become so complex that these outliers just shouldn’t exist. This is in no way a complaint about the new generation of stats. I love them. I love that the history of baseball is the statistical record. What I don’t like is when stats are used to manipulate the reader into dismissing great performances. What Jimenez has done to start 2010 has been simply amazing. For comparison’s stake let’s look at how Jimenez’s numbers compared to 1968 Gibson’s season and 1986 Clemens’ season. The first graph shows BABIP

Data would suggest that Jimenez was on par with the big boys at least through 10 games. Since then it would appear Jimenez is trending back to the mean. Should be interesting to see where Ubaldo goes from here. He did have that rough start in the rain and during the next stretch of games against the leagues best ought to determine whether this is something to remember. I just want to see 20 wins! But why be a downer let's focus on the goo things

His numbers - Jimenez is 13 -1. He has pitched 101.1 innings. He has allowed 65 hits and 13 runs. He has struck out 88 and allowed 36 walks. He has averaged 7.2 innings an outing and has an average game score of 69. Batters are hitting 0.189 against him. Of the 385 batters he has faced only 56 have gotten to a full count. He has faced 75 batters with runners in scoring position and they are batting 0.147. The telling stat for the home team is that he has won 13 of the 36 Rockies victories and ten of wins have come after Rockies losses. Regardless of any stat a pitcher’s job is to put his team in the position to win. How the pitcher gets there is some crafty pitching, some luck, and timely hitting by your side. Baseball is a long season and time will tell whether these numbers will hold up. I think Jimenez will probably hit a rough patch in July and August. The team behind him is in disarray. Scoring runs has been the Rockies Achilles heel not to mention an on and off again bullpen. His innings pitched has raised a few eyebrows but I also think that when he is on the mound the players behind him step up. In an unmeasurable stat I do think a single player’s performance can bring out the best in his team mates so who knows what might happen. Heralding a particular player at this point in the season as the greatest is a bit premature. At this point though I know I would take Jimenez as my starter is I needed to win one game.

Monday, June 14, 2010

Rox Talk - Week 10

The Week That Was
A 4-3 week against the Astros and Blue Jays. The Rox currently stand at 33-30. Four games back from division leading San Diego in fourth place. Currently 19-12 at home and 14-18 on the road, the Rox have outscored its opponents 291-244 (expected wins is 37 versus historical wins at 29). On pace for 85 wins with 748 runs scored and 627 runs against. Playoff indicator (Runs scored/Runs Against ratio) is 1.19 (>1.16 indicates high playoff potential).

Well the interleague record continues to be awesome but we lose 3 of 4 to the worst team in the NL...go figure. That is baseball. Tracy keeps his streak alive...that is winning home stands. Now if we could only do that on the road. Wacky Denver weather this years continues to play havoc with us fans. I am sure the nightmare conditions would have been called earlier on Friday night if it hadn't been an interleague game. Have to feel sorry for Ubaldo Jimenez and giving up a couple of runs especially with a rainy night and wet ball. Either way it was still impressive that he got win number 12 under those conditions. When this season is over it will be interesting to see if the year of the pitcher survived for 162 games or if the first third of the year was a fluke. Will Jimenez be the symbol of the season like Bob Gibson's stellar season in 1968? Let's hope they don't decide to drop the mound another 5 inches.

Win Probability Added
I love win probability added. I discovered the stat back when I read Alan Schwarz's "The Numbers Game" I have been studying it ever since. It is a pretty esoteric stat but to me it boils the game down to batter vs hitter vs historical probability. Sure the pundits will say that WPA is meaningless that it doesn't truly present what a player can or statistically predict but to me it represents the beauty of baseball. While all events in a game are random, the underlying probability suggests that most baseball outcomes are very predictable, but every once in a while something magical happens...a hitter gets that base hit or a pitcher gets that strikeout. The big stats just sort of get muddled because at the end of season someone has 40 home runs but do you remember any of those or do you remember that ninth inning dinger that sent you home happy?

Last week the big news with the Rox was the inability for the offense to hit with Runner's in Scoring Position. Data would also suggest the Rox aren't scoring a lot of runs after the fifth inning (203 runs before the fifth versus 84 after the fifth through the ninth, compare this to 2009 when 433 vs 361 runs). Is our offense this maligned?

Another way to look at it is through the lens of WPA. Two components (offense vs pitching/defense) contribute to your chances of winning. Winning teams components equal 0.5 and losing team components equal -0.5. Baseball Reference conviently tracks these components so you can look at which component contributed the most to the Rox win. Of the 63 games played in 2010, the pitching component has been greater than the offense component 37 times. Of the 33 wins the pitching component has been greater than the offense 21 times. So in the bulk of the Rox wins the pitching has been the reason. In the 30 losses though, the components breaks out 14 vs 16 in pitching and offense (the component contributing the least to the effort) which at least suggests to me that we can't blame the hitting on everything! So the figure below shows each of the Rox game plotting the hitter and pitching WPA per game. Top data points in purple are the wins and bottom points in grey are the losses. Red dots indicate the .25 mark which would suggest an equal effort on the part of the offense and the pitching/defense.

Guess the thing that jumps out to me first is the fact that we haven't won a game without a positive pitching WPA. Although to have this happen you'd probably have to have a ninth inning meltdown by the staff followed by a two out walk off homer...Rox haven't had too many of those. It also seems the wins are more tightly grouped versus the losses. Will have to revisit this after the end of the season. Below is just another way of looking at the data plotting individual pitching or offense WPA with wins.

NY Times Keeping Score
Beautiful graph by NY the simplicity of it.

Monday, June 7, 2010

Rox Talk - Week 9

The Week That Was
A 3-3 week against the Giants and D-Backs. The Rox currently stand at 29-27. Four games back from division leading San Diego in fourth place. Currently 15-9 at home and 14-18 on the road, the Rox have outscored its opponents 261-222 (expected wins is 32 versus historical wins at 25). On pace for 84 wins with 755 runs scored and 642 runs against. Playoff indicator (Runs scored/Runs Against ratio) is 1.18 (>1.16 indicates high playoff potential).

The next 3 weeks will determine the season. 13 of the next 16 games are at home. Hopefully after some good home cooking the Rox will find themselves in the lead or a game back. Things to look for: Starting pitching settling into a routine, bullpen maintaining its dominance, winning some close games, and a more stable lineup without the juggling. I think these are the keys.

A lot was said over the weekend about the lack of hitting with runners in scoring position. Spreadsheet below shows certain offensive statistics with RISP over the last few years
Nothing overtly earth shattering. Two things of note at least to me is that if you take the ratio of runs to hits, the Rox, when hitting (!), are scoring. The problem is the hitting is down (obviously). Another highlight is the number of double is waaaayyyy down. Average number of doubles hit with RISP is around 83...Rox are on pace this year to only hit 55 (can you say Helton!). Of course Rox have played a lot of games on the road where their doubles tend to be lower. Also now that Helton has contacts maybe he will be able to see the ball...any why do the players not all have eye exams each year in Spring Training? To me that would be something pretty obvious!

Finally one note of concern is the lack of run scoring or production after the 5th inning. Historically the Rox have scored 55% of their runs in the first 5 innings and 45% over the last 4. This year through 56 games Rox are scoring 70% in the first 5 and 30% in the4 last 4. Rox are either bashing their way to an early rout or going to sleep in the later innings...big time!

Third Inning
A 162 game schedule can easily be broken into "9 innings" of 18 games a piece. Over the years I have tracked monthly averages and I am always amazed how similar the averages are. So this year instead of months I am simply tracking 18 game stretches to see if anything changes. The spreadsheet below shows the first, second, and third inning.

Amazing how similar stretches of games are and yet games themselves are so vastly different. Goes to show how statistics can mess with you!