Friday, June 29, 2018

The Results of the 2016 Presidential Election Using The Proportional Electoral College

As I'm sure you all remember, six years ago I fixed the Electoral College. It's pretty simple, instead of awarding states as winner take all, electoral votes are awarded proportionally to the candidates based on what percentage of votes in the state each candidate receives. There are a lot of benefits to this plan:

1. The biggest advantage is the perception that every vote matters. My whole life, elections have focused exclusively on swing states. States like California, Texas and Illinois are ignored because it's a foregone conclusion. Not in this plan. Illinois has 20 electoral votes and they might go 12/8 or 14/6 or 10/10. So there are no wasted votes. Once people realize this, I would expect turnout to go up.

2. Likewise, using the a national popular vote creates the impression that a single vote doesn't matter in a sea of 130 million votes. Also, a national recount would be a disaster.

3. It just kind of makes sense. If we were coming up with a system today, would anyone propose that states should be winner take all. Does it make sense that if you carry a state 49 to 48 percent of the vote that you should get credit for winning all of the state? Proportional makes every vote count. Blowouts matter. If you get 90% of a state you should be rewarded more than your opponent who got 51% of a different state.

4. It actually recognizes that 3rd-party candidates got votes. If we actually did this, instead of requiring 270 to win, I'd just say the winner is who has the most electoral votes as 3rd party candidates make it difficult to get to 270 in a close race.

5. The biggest cultural impact is eliminating the idea of red and blue states. Not sure if you've noticed, but party lines are toxic and I think it would help quite a bit to realize that no states are entirely red or blue, that we're all mixed together, living side by side.

All that said, how would the 2016 election have turned out with this system in place?

(Methodology: I first multiple a candidates percentage of the vote times the number of electoral votes and round to the nearest whole number. 6.48 becomes 6. Do this for all candidates. If there are any leftover electoral votes(this is common), I award them to the candidate that won the state. If there is an extra (this is quite rare) I would deduct from the candidate that won the least votes. In the cases of Hawaii, Texas and Washington, there were faithless electors so Trump + Clinton's actual total is only 531 electoral votes, but I'm awarding all 538.)

Click here to view the full spreadsheet with 3rd-party candidates.



How is this possible? How does an election with winner take all go from Trump 304 - Clinton 227 to Clinton 269 - Trump 260?

Clinton racked up bigger wins in her states. She gets credit for carrying California by a 2 to 1 margin and only barely losing Texas. If you ignore the names involved here, doesn't that make more sense. Shouldn't you get credit for all your votes? It would be like a baseball team giving up 8 runs in the 4th inning and 9 runs in the 7th inning but claiming victory because they had a better score in 6 out of the 9 innings.

Another nice thing about this system, is that it inherently doesn't favor one side over the other. It's not prone to pumping up the Democrat or punishing the Republican. It's simply making every vote count. In a 2020 election, it's just as likely that this system would help a Republican. In 2012, it would have portrayed a much closer election than the Obama 332 - Romney 206 result.

Tuesday, June 26, 2018

World Cup 2026

So the World Cup is coming to North America in 2026. Great!
They're expanding from 32 to 48 teams to be more inclusive. Great!
They are going to have 16 groups of 3 teams each, 2 group matches instead of 3 for every team. Oh no.

- - -

3 teams in 16 groups, where the top 2 teams advance to a 32-team knockout round?

The first thing this instantly robs fans of is the simultaneous matches to conclude the group stage.

Let's look at a scenario involving a fictional Group with Brazil, Croatia and Tunisia.

Brazil play Croatia first and win. Now it's Croatia vs Tunisia. If you're Tunisia, you know that two draws will get you in. So your best shot is to play super conservative against both Croatia and Brazil, keep 11 guys back and allow zero goals. That's not great soccer or fair or good for anyone.

- - -

Here's another group: Spain, Australia, Senegal.

Spain beats Australia. Senegal and Australia tie. Now it's Spain vs Senegal. They both know that a tie is great for them. Spain gets to win the group, Senegal advances. So they both play conservative and it ends 0-0. That seems super realistic, right?

- - -

So 16 groups of 3 teams is shit. What's the alternative? The obvious is 12 groups of 4.

The top 2 teams advance, that's 24. Then you sort the 3rd place teams and the 8 teams with the most points advance. If that seems familiar, it's what the World Cup 1986 to 1994 when they used the group stage to narrow 24 teams down to 16.

But leagues and clubs are concerned with the total number of matches. So instead of fairness, we're left with money determining the result. Sounds like FIFA after all.

Saturday, June 23, 2018

World Cup Math

In 4-team groups, there are 6 matches and a maximum of 18 points, with a minimum of 12 points, that can be earned.

It's impossible for two teams in the same group to each finish with 9 points each. (In 2002, Brazil won all three group matches and all four knockout matches to win the Cup.)

It is possible for three teams to finish with 6 points each, which means a team could get first place with 6 points while another team gets eliminated with 6 points. (In 1994, there were two instances where three teams finished with 6 points each, but because there were only 24 teams with 16 advancing, all teams with 6 points advanced.)

It is possible for three teams to finish with 5 points each, if three teams win one game and draw the other two. This has not happened as far as I know.


It is possible for all four teams to finish with 4 points each, if each team gets exactly 1 win, 1 draw and 1 tie. This happened in 1994. It was the only time every team had the same points and same goal difference.

It is possible for all four teams to finish with 3 points each, by tying each game. This has never happened.





Monday, June 18, 2018

Scoring in Soccer

The very first live post on this site was a bunch of stupid suggestions under the title, How to fix soccer. I rightfully got a scolding from pre-twitter soccer fans. I'll hopefully do better with this post.

- - -

Every World Cup since 1962 has averaged between 2 and 3 goals per match. (The 2010 and 2014 World Cups averaged a combined 2.47 goals per match. That's 1.23 per team per match.)

In 1958, 3.6 goals per match.
In 1954, 5.38 goals per match.
From 1934 to 1950, every World cup averaged between 4 and 5 goals per match.

- - -

I'm not saying that the game is broken or needs fixing. But it's worth noting that scoring is half of what it's used to be. Goals are exciting and important. In doing research for this, I found this from a Slate article:


I agree that fans want close games, of course, but I also think fans like scoring. Not too much, mind you. For example, Portugal just played Spain to a 3-3 draw and I think most people found it very exciting. 12-9 would have been too much, but 3-3 is great. 5-4 is fun every now and again. 

In the NHL this year, teams averaged 2.97 goals per game. That's per team, so we're talking 6 goals per game. Some games you get 1-0 nailbiters, some are 7-6 barnburners and some are 5-0 blowouts. But six goals a game works nicely. The arena explodes every time there is a goal. At 6 per game, the rarity does not diminish the excitement. (A good example is a NBA basket. There are 40 made per team per game. So you're not exploding out of your seat or crying if your opponent makes a shot. Likewise in tennis.) 

In the NFL, teams average about 2.5 TDs per game. Same story as hockey. Every TD matters a lot. Games are fun when there are 4 and 5 touchdowns per team, and still fun for the rare 14-7 game too. 

So for soccer teams to be averaging 1.3 goals per game...it's a little weak. Especially when you throw in that there are straight draws after 90 minutes. The possibility of a 0-0 draw is thoroughly present. 

I think the numbers show that increasing from 2.5 goals to around 5 goals per game would be healthy for the game. And it would actually keep it in line with its historical roots. 

- - -

And one more thing...the Penalty. 

I tried to find out how many goals are typically scored via penalty. Hard to find, but I found a video showing "all 12 penalty goals from world cup 2014". So if you take out the 12 goals scored via penalty, the average for 2014 drops from 2.67 to 2.48 per match. Goals via penalty account for 7% of all goals at the last World Cup. 

Here's the thing. Goals scored in the run of play are beautiful and exciting and either well-earned or a huge mistake from the opponent. Goals scored via penalty are not beautiful, not that exciting because 76% are converted and often come down to a judgement call—where well-earned or mistakes may not even come into it. 

It also often just doesn't make sense. A little push or trip that five feet away would be given a difficult contested shot from 25 yards is now given a gimmie uncontested shot from 12 yards. Combined with the overall low-scoring nature of the current game, one judgement call can decide the whole game. 

So if I was in charge, which I'm not, I would try to increase scoring in the run of play and decrease the impact of goals via penalty. 

- - -

Let's start with the penalty. In the NFL, there used to be two kinds of face mask penalty. A severe penalty for intentionally grabbing the face mask and a milder penalty for accidentally grabbing the face mask. It worked pretty well. Let's bring that idea to soccer. Instead of the current 12-yard spot, let's have 2 spots. One from 10-yards for obvious, blatant penalties (intentional handballs, reckless slide tackles) and one from 16-yards for more inadvertent fouls (unintentional handballs, pushes and trips that were attempts on the ball but missed). 

The exact distances would warrant testing but the goal of this is simple. The odds of converting a penalty should be commensurate with the odds of scoring on the play when the infraction occurred. If a defender wipes out a player on a breakaway, the offense should be given a kick with 90% conversion rate. If the defender was going for the ball in a normal offensive set, the offense should be given a kick with a 50% conversion rate. 

This 16-yard spot would become more exciting as goalies would be able to wait longer before diving. It's a more challenging kick and would have more inherent drama on each kick. 

- - -

As for increasing scoring in the run of play...there are lots of possible solutions. The obvious is to increase the size of the goal, even by just a foot on each side. It's currently 24 feet. A 25 or 26 foot goal might be perfect. 

The game is good and doesn't need radical changes. But any game warrants looking at ways to make it better. 

Tuesday, June 12, 2018

IHOB

It's not very often that an advertising campaign actually gets people talking so I thought I'd weigh in on IHOB. Without actually knowing any inside information, I'm 99% sure I know what happened.

It starts with someone on the client side not even in marketing. They're looking at their numbers and they can see that they're doing okay at breakfast time, but slow down at lunch and dinner. Makes sense, of course. But they want more revenue so they tell their Marketing department to get more people in for lunch and dinner.

Marketing looks at the menu for a minute, and decides why not burgers? They augment this decision with charts of burger sales and think that if we can increase our burger sales even by 1-2%, that's a successful campaign. So they get approval from their bosses. They start working on new burger recipes and then tell their Agency: We need an advertising campaign to sell more burgers.

The creative team goes to a meeting and is told to come up with a campaign to sell burgers. There's a lot of eye-rolling but it is what it is. They were most likely told it's a small budget but they want to get people buzzing.

Creatives go off and come up with a bunch of ideas and this one probably started as a sarcastic joke. Why don't we just change the name to International House of Burgers. And then their partner was like, wait, we could just flip the p to a b. So they present it and everyone agrees this would be a bold thing that could get people buzzing about their burgers. It's a stunt that hasn't been done. And remember, their goal was to get people buzzing about their burgers.

So it all gets approved and they start with a teaser of We're changing our name. The public sees this and starts buzzing. This is bold. People are interested.

And then yesterday, IHOB is revealed. People are roasting them. Because it's really dumb to go to IHOP for burgers.

The creative team did their job. The marketing team on the client side had good intentions. So where did it all go wrong?

They came with a great solution to the wrong problem. The problem is you want more sales. And yes, it's better if they come at lunch and dinner. The problem is not "we're not selling enough burgers." 

If they had come to the creative team with a blank canvas: "let's change our image" so more people come in, that could have worked. Maybe they'd change their meaning to some other interpretation of IHOP, maybe they change it to International House of Brunch, or maybe they come up with food solutions, lean into the international side perhaps. There's lots of solutions to that problem. 

- - -

At they very least, it would make sense if their burgers actually were international. But they're not. Same flavors you've been seeing at Chili's and Red Robin your whole life. So yeah....

- - -

I think International House of Brunch would have been really smart. It's still on-brand, but could appeal to younger audiences, and would have solved the lunch/dinner issue too.