In light of the Quebec election yesterday, I’d like to elaborate on a similarity that I’ve seen between baseball and the Canadian electoral system.
The final results from the election are in and I know a lot of Americans might be perplexed by the way that the numbers break down. Out of 125 seats in the Quebec National Assembly, the Liberal party won 70 seats (56% of the seats which is more than enough to form a majority government) while the Parti Quebecois took 30 seats (24% of the seats) and the Coalition Avenir Quebec (CAQ, the Coalition for the Future of Quebec) took 22 seats (17.6%) and the Quebec Solidaire party took three seats (2.4%).
Simple enough, right? But wait! Let’s look at the percentage of the votes that each party got. The Liberals had 41.5% of the vote, the PQ had 25.4%, the CAQ had 23.1%, the QS had 7.6% and other candidates had 2.4% of the vote.
Wait, what? For those of you unacquainted with our Canadian electoral system, we use something called “First Past the Post” (FPTP) that we inherited from the British. In this system, the country (or, in this case, the province) is divided up into regions called ridings. Each riding elects a representative. The candidate in each riding who wins the most votes wins the seat. The leader of the party that wins the most seats becomes the Prime Minister of the country or the Premiere of a province. This is what enables a party that has less than 50% of the vote to actually win a majority of the seats in the legislature, especially when there are more than two parties.
Translating this system to baseball, runs (the aggregate of runs scored and runs allowed) are like votes and wins are like seats in the legislature. In the Canadian electoral system, the goal is not to get votes, it’s to win seats. The same way that the goal in baseball isn’t necessarily to score runs, it’s to win games no matter how you have to do it. In politics, to win of a seat by just one vote is the same result as a win by 3,000 votes and in baseball, a win by one run is the same as a win by 10.
For me, this has a huge impact in the baseball writing world. While many of us (myself included) extoll the virtues of advanced stats that more accurately measure a player’s performance over time, we can’t overlook the fact that most teams don’t look at a player’s quality in quite the same way. While advanced stats may evaluate a player’s overall performance in terms of runs produced or, going one more level up, in wins produced (stats like WAR assign a win for every 10 runs that a player produces/saves), they don’t directly correlate to a team’s ability to win games.
Why not? Because a team’s ability to win games is determined by how the players perform in specific, real world contexts. Advanced stats are (preferably) built over contextless, large sample sizes that allow the analyst to filter out any statistical noise like park factors and luck. These large sample sizes are great for analysis of a player’s overall ability but don’t really quantify his ability to contribute to his team’s win-loss record.
Just as an election can hang in the balance of just one riding and that riding can be decided by a single vote, a full season can depend on the outcome of a single game and that single game can depend on the outcome of a single at bat for which context and luck are paramount. The variability of a player’s performance in just one at bat is almost limitless. If you look at a whole season’s worth of data, a player will have had a precedent for just about any possible outcome and the potential always exists for a new outcome to occur. Sure, in a decisive situation, Miguel Cabrera is likelier to produce a more positive result than Adeiny Hechavarria but Hechavarria has the potential to contribute as much or more directly to his team’s wins and losses than Cabrera by virtue of performance when the game hangs in the balance. The Win Probability Added (WPA) stat has the ability to reflect this tendency. In 2013, Paul Goldschmidt had a much higher WPA than Carlos Gomez despite a lower WAR (based on Baseball Reference Wins Above Replacement and Win Probability Added).
This is why advanced stats, while useful, can’t be heralded as the answer to every question about baseball, especially when it comes to figuring out how a team actually plays in a full season. Things like Pythagorean record are estimates based on how many runs a team scores and allows. These runs, like the number of votes a party gets over an entire election in Canada, can give you an idea of what the result might be but the real world, game-by-game results can differ quite drastically. Much like an election, you have to look at each game one by one wherein a team can seriously outperform expectations by good situational hitting, pitching and defense. Where one play can turn an expected loss into a win (or vice versa), in politics, a single vote could mean the difference between winning and losing a particularly riding. In elections in Canada, only the seat totals matter, much like in baseball, it’s only the final win-loss record that matters.
By compiling several of these close victories, teams and parties can drastically skew the results of a season without large margins of victories that make the popular vote totals (or run differentials) more convincing. The 2012 Baltimore Orioles are a perfect example of this phenomenon, compiling a 29-9 record in one-run games and outperforming their 82-80 Pythagorean record by 11 games, finishing 93-69 despite scoring 712 runs and allowing 705.
Situational hitting, productive outs, timely defense and big strikeouts do matter in professional baseball. When calculating wOBA or xFIP, individual stats are all weighted equally (e.g., for wOBA, all singles are equal and for xFIP, all strikeouts are equal) but in the context of an individual game, they are definitely not all equally important. While advanced stats are very valid to understand what a particular player is likely to do, they can’t predict what a player is going to do in a particular situation at all. Using the model of the Canadian electoral system, we have a way of understanding the context-based necessity of playing the games in the real world.
Don’t forget about The 2014 Toronto Blue Jays Minor League Handbook, now available as an ebook at Smashwords.com for $7.99 US. It’s coming soon to Amazon, Apple iBooks, Kobo and other fine retailers. You can purchase and preview the book at our Smashwords.com page! If you like us here,“like” us on Facebook!