BrewCity...

[Gato78]

I have to ignore metrics at a certain point. Baseball guys in particular are way too deep into metrics (basic: left handed pitchers fare better against left handed hitters--time immemorial). I want gamers, guys who can hit free throws or get to the rim when the game is on the line or even when a big basket is needed. I do not give a damn what their e%FT/assist/in January rates are because there is an immeasurable. The guy who shoots 40% may be the guy who has the stones to lead a rally when 10 points down--even if he misses some shots. That toughness cannot be measured. If metrics ruled we would never have beaten Kentucky in 2003--or 1994 for that matter.

[TheSultan]

I have to ignore metrics at a certain point. Baseball guys in particular are way too deep into metrics (basic: left handed pitchers fare better against left handed hitters--time immemorial). I want gamers, guys who can hit free throws or get to the rim when the game is on the line or even when a big basket is needed. I do not give a damn what their e%FT/assist/in January rates are because there is an immeasurable. The guy who shoots 40% may be the guy who has the stones to lead a rally when 10 points down--even if he misses some shots. That toughness cannot be measured. If metrics ruled we would never have beaten Kentucky in 2003--or 1994 for that matter.

Or the 2013 team. I think we are going to look back on that team (BE champs, Elite 8) as one of the more improbable runs in recent Marquette history. Even though they weren't a great outside shooting team, they could get to the rim, get to the line and rebound pretty well. And defend. Not a single NBA player on that team too.

[Nukem2]

[QUOTE=TheSultan;134169]Or the 2013 team. I think we are going to look back on that team (BE champs, Elite 8) as one of the more improbable runs in recent Marquette history. Even though they weren't a great outside shooting team, they could get to the rim, get to the line and rebound pretty well. And defend. [B]Not a single NBA player on that team too[/B].[/QUOTE]

Not to be picky here, but Vander has played 5 games in the NBA.

[ValiantSailor]

How many games did "Hack-a-Shaq" win? More than one, I'll bet.

VS

[TheSultan]

How many games did "Hack-a-Shaq" win? More than one, I'll bet.

VS

Nate Silver did a statistical analysis on the Hack a Jordan strategy and shows that it actually increases the expected net points scored by the Clippers.

http://fivethirtyeight.com/features/...dan-is-futile/


And while this isn't in-depth statistically, it shows that Hack a Shaq was largely ineffective.

http://www.espn.com/blog/truehoop/po...-shaq-strategy

From 2000 to 2006, Shaq’s teams won 22 out of 25 playoff series. Two of the three series he lost were to teams coached by Larry Brown, whose “play the right way” mantra prevented him from fouling Shaq without the ball every time down the floor.

“I’m not going to make a farce of the game,” Brown once said. “I’d rather have people beat us in basketball plays.”

...

Perhaps I could endorse the constant stoppage of play if fouling bad free throw shooters was the way to beat them. But with Shaq it was actually the opposite. At his peak, the more free throws he shot the better his winning percentage. From 2000 to 2006 his teams were 22-4 in playoff games when he attempted 15 free throws or more, including 6-1 when he attempted at least 20 free throws.

[Alan Bykowski, "brewcity77"]

I have to ignore metrics at a certain point. Baseball guys in particular are way too deep into metrics (basic: left handed pitchers fare better against left handed hitters--time immemorial). I want gamers, guys who can hit free throws or get to the rim when the game is on the line or even when a big basket is needed. I do not give a damn what their e%FT/assist/in January rates are because there is an immeasurable. The guy who shoots 40% may be the guy who has the stones to lead a rally when 10 points down--even if he misses some shots. That toughness cannot be measured. If metrics ruled we would never have beaten Kentucky in 2003--or 1994 for that matter.

Metrics are not there to determine a one-off game. Yes, they can help determine outcomes, but the story they tell is the bigger picture. That said, the metrics in 2003 say we had the most efficient offense in the country. Marquette was top-10 in 3P% and top-25 in eFG%. Was it still an upset? Yeah. Would Kentucky have beaten us 7/10 times? Most likely. But metrics are about probability and give you a sketch of what could happen in a given game, not a definitive answer.

On the other hand, they tell the story of a season far more accurately. There are obviously outliers, but I'd guess that if we took Pomeroy's predictor right now and ran it out to the end of conference play, most teams would be within 2 wins of his current predictions. Because metrics are better at telling a big picture story than a snapshot.

Last year, I ran a similar test using the RPI Wizard application on RPIForecast.com. I tested around 25-30 teams and found that the Wizard predicted RPI accurately within 5 spots over 54% of the time and within 10 spots over 78% of the time. And that was strictly using aggregate win totals, not looking at how accurate it was when factoring home W/L outcomes.

You can throw it out, but this stuff works. No, an advanced metric won't determine if Howard will make the corner three with 1.2 seconds on the clock while trailing 78-76 at Cintas, but they'll do a good job of telling you how good your team actually is and how they match up against given foes.

[MUAlphaBangura]

Metrics are not there to determine a one-off game. Yes, they can help determine outcomes, but the story they tell is the bigger picture. That said, the metrics in 2003 say we had the most efficient offense in the country. Marquette was top-10 in 3P% and top-25 in eFG%. Was it still an upset? Yeah. Would Kentucky have beaten us 7/10 times? Most likely. But metrics are about probability and give you a sketch of what could happen in a given game, not a definitive answer.

On the other hand, they tell the story of a season far more accurately. There are obviously outliers, but I'd guess that if we took Pomeroy's predictor right now and ran it out to the end of conference play, most teams would be within 2 wins of his current predictions. Because metrics are better at telling a big picture story than a snapshot.

Last year, I ran a similar test using the RPI Wizard application on RPIForecast.com. I tested around 25-30 teams and found that the Wizard predicted RPI accurately within 5 spots over 54% of the time and within 10 spots over 78% of the time. And that was strictly using aggregate win totals, not looking at how accurate it was when factoring home W/L outcomes.

You can throw it out, but this stuff works. No, an advanced metric won't determine if Howard will make the corner three with 1.2 seconds on the clock while trailing 78-76 at Cintas, but they'll do a good job of telling you how good your team actually is and how they match up against given foes.

And then they play the games and all that stuff means absolutely nothing.

[Alan Bykowski, "brewcity77"]

[/B]

And then they play the games and all that stuff means absolutely nothing.

That's just idiotic and ignorant. Stats tell us Marquette is a good three point shooting team. Is that meaningless? Stats tell us they're a mediocre defensive team. Does that go out the window? Stats tell you what you are.

Teams can either learn from them, use them, and improve with them, or ignore them and stagnate. Fans can do the same. Realize that the eye test isn't perfect and only tells you so much, or accept that this stuff has value.

There's a reason teams like Golden State and Cleveland have separated themselves from the pack. Because they use and improve with advanced metrics. You can embrace them or be left behind. Saying they mean absolutely nothing sounds exactly as intelligent as saying the brewcityball rec team could beat the Bucks, because hey, all those stats mean absolutely nothing!

[Gato78]

But....Brewcityall may have better metrics than the Bucks, but you know, because of the eye test, the Bucks will win. Brewcityball may have phenomenal 3pt shooting but the stat means nothing until they actually play the Bucks. When teams are fairly equal, one may have a better percentage of this or that that gives them the edge, but I will take the team with guts over the team with a better eFG% every day of the week. Metrics only go so far.

[TheSultan]

This is a topic about stats, and what stats are considered more important and why. Therefore, I don't understand why people come into the topic and say things like "metrics only go so far."

No kidding. No one claimed otherwise. No one claimed that the games were played on a computer and that the team with the better eFG% was going to win every game.

And I am curious Gato, what a team with "guts" looks like in your mind? What do they do that shows they have "guts?"