THE BOOK--Playing The Percentages In Baseball

I like it.

Another thought I once had to help eliminate some clock issues is this: when the buzzer goes, the game doesn’t end.  It continues until either:

a) the team behind fails to score on a possession; or

b) the team behind fails to prevent the other team from scoring on a possession.

That is, even though you’re behind, you get to keep playing until you make a mistake either on offense or defense.  The benefit is that it’s not as important to waste the clock.

Having said that ... I like this proposal better than mine.

Naively, you’d expect the second “half” to take less time on average than the first—the high-scoring team has fewer points to score than they already did. I don’t know if strategy shifts would compensate for this or not, but it might mean you’d want the timed segment to run 25/30 minutes instead of 20/24, or something.

I just thought of something.

1. In baseball, a team that has fewer hits+walks than its opponent can win often enough.  Say if one team gets on base 10 times and its opponent gets on base 15 times, the lower team might win say one-third or one-fourth its games (or whatever it is).

2. Same situation in tennis.

We know in basketball, there is much less randomness in game outcomes than baseball.

So, how about we introduce some randomness by following a game-set-match kind of scoring system, like tennis?

Someone can sim a few games for us, but you can make it that a game is 6 points (and win by 2).  And you have first to win 3 games per set (no win by 2). And best 3 of 5 sets.

(We can work out more realistic numbers.)

So, the average game might be 8-4 (I expect alot of tie-breakers), the average games per set would be 3-1, and you’d end up with an average of 4 sets.

So, total points would be 12 x 4 x 4 = 192

You can have a shot clock, but no game clock.

What do we think?

I could imagine scenarios where fouling would now start to occur at the end of the 1st half. Say a team is down 40-20 with a couple minutes left. It would be a big benefit to get to the half with the score like 55-35 or something like that. Even if you don’t cut into the lead you’d give yourself a much better shot at coming back in the 2nd half. I guess this would be lessened by the fact that you’d have to worry about your players fouling out early in the 2nd half, so it might not be a big concern.

I do think a solution like this would be very helpful for college basketball as a lot of the games are just ridiculous at the end. An easier solution might just be cutting the shot clock down by 5 seconds. IMO the amount of time fouling happens in the NBA seems to be a lot less. The shorter shot clock is one reason. Of course they have a lot more games (regular season and the playoffs) while in college every loss seems to be critical.

Is there still a shot clock? That would ensure scoring stays somewhat high. Especially if the shot clock time was lowered from 35 seconds to 25 or even 20, for the 1st half. Nobody could dilly dally, they’re HAVE to shoot or steal the ball every 20-25 seconds. Steals aren’t going to happen enough to keep the score down too low, ‘cause players will guard against that a lot better once everybody is trying to steal it from them.

Why not just have it be first team to score X number of points wins?  First team to 100 wins...with quarters ending when a team hits 25, 50, and 75 points.  Or is that just too simple?

First team to 100 and up by four or more points wins.

I like the 100 point thing. 

I think what Nick was trying to do when he proposed his is that each league/level would have its own point system.

This is especially useful in high school, when teams are lopsided.

So, I can see how fixed points works in NBA, but not sure about the other level.

Why do we even have halves?  Are basketball teams really in dire need of a 15 or 20 minute break?  Never played higher than high school, but it always felt like halftime was needless.  Just a time to sit around and b.s. and listen to the coach yell at ya or make hairbrained adjustments we’d all ignore or forget 2 minutes later.  Same is true in football. 

Just play first to 100, gotta win by at least 4.

Half-time = commercials.

In the old one-on-one tournaments they played back in the 70’s the format was play to 20; had to win by four.  I remember the format working well.  Games would always end on a basket by the winning team.  That would be exciting.  However, games between Washington and New Jersey would probably have to start at 4:00 in the afternoon because people would have to get up the next morning for work.

A clock does make the game a bit easier to package for television - less uncertainty over how long a game takes.

While a playing to reach fixed point total (e.g. 100) has the advantage of simplicity, that doesn’t scale well across different levels, or even team styles within a level. I don’t immediately have older data, but the current year NBA scoring average is 95.45, down from 99.55 last season. So playing to 100 would mean about a 4% longer game this year, and variation across teams would be even wider.

Now this year the lockout may well explain the bulk of the difference, but variations in style of play can also affect point totals. Also I’d note teams play overall tougher defense in the playoffs (last year the post-season scoring average was 94.02). And league-wide scoring averages have ranged from the low 90s to far above 100 in different years.

Playing to some point total, even a dynamically determined one (as in the original post) does force the winning team to keep trying to score, whereas now with a big enough lead a team places more emphasis on running down the shot clock each possession. Such a change would likely greatly reduce “garbage time”, when coaches put scrubs in for a few minutes late because the outcome is assured.

Thanks to Tangotiger for sparking some discussion about my idea! I’d love to find a way to ensure that basketball teams always play by the spirit of the rules and adhere to the basic objectives of the game.

For what it’s worth, I envision a slightly different approach from what’s expressed in the original post in the way that this new format would prevent an undesirably short second half. For most games, the teams’ first half scores would simply be added together to create the “victory number” for that game. However, a minimum number of points (not any kind of percentage) would need to be enforced in games where at least one team has a low-scoring first half (it would be added to the leading team’s halftime score to create the victory number). I think 25 would make for an appropriate minimum number for NCAA DI men’s basketball.

Examples: If the halftime score is 35-31, play to 66 (minimum not enforced); if the halftime score is 30-23, play to 55 (minimum enforced); if the halftime score is 22-19, play to 47 (minimum enforced)

Thanks and keep the feedback coming!
Nick

My biggest concern about the timeless format (as I call it) is whether teams will be truly motivated to score during the first half. I’ve thought of it every which way, and I’m convinced that teams should always try to score and always prevent their opponent from scoring (that’s the idea, right?), but I think some coaches/teams will go out of their way to keep first-half scoring low (however misguided that strategy might be).

Even this wouldn’t be the worst thing in the world - currently, we hear all the time in college basketball about how one team wants to play a halfcourt style and keep the score in the 50s or 60s, while their run-and-gun opponent wants the score in the 70s or 80s. I don’t think there’s any harm in that.

I think in the timeless format, we might see underdogs employ a slow-it-down strategy in the first half, hoping to set up a sprint to the finish in the second half that might lead to a fluky win (I’m not sold that this is the best strategy). Favored teams would (and SHOULD) try to build up a big first half lead to avoid the previous scenario.

In response to post #1: I considered all different kinds of formats, but I hadn’t considered the idea you presented. I really like it, and I think it’s a brilliant way to reduce (or even eliminate) the practice of deliberate and repeated fouling (playing foulball, as I call it) we see near the end of games currently. I don’t think it would be as effective in preventing the leading team from stalling (playing stallball).

...to post #2: this isn’t naive at all! You’re absolutely right - if teams play the same way they did during the first half, the second half would be shorter, and that would seem to be a drawback of this format. I have an idea toward the same end you mention (to compensate by lengthening the first half) - during the first half, stop the game clock (don’t stop play, though) after each made field goal and restart it after the ensuing inbounds pass. It doesn’t seem like much, but doing so would add about 1:15 to the first half (based on data from the 2007 NCAA tournament, when I was really researching this idea hardcore) and a theoretical 1:15 to the second half.

...to post #4: You might be right (but I hope you aren’t!) - teams might begin fouling late in the first half. But two considerations put my mind at ease about this unsavory strategy:
1. As you mentioned, a trailing team would compound their problems by putting their own players in foul trouble.
2. A 55-35 deficit is probably more appealing than a 40-20 deficit. But while playing foulball is a good way for the trailing team to bump up the combined score, it’s also a good way to widen their deficit (at least it did, overwhelmingly, during the 2007 NCAA tournament). Consequently, I don’t think trailing teams would play foulball during the first half.

...to #5: Yes, there would be a shot clock throughout the game.

...to #6-8: Playing to 100 (or any number chosen before the start of the game) is the first idea that comes to mind when trying to avoid the negative side effects of the game clock, but the actual length of games would vary way more than I (or any television executive, as post #12 suggests!) would like to see (also, the win-by-four requirement eliminates the prospect of true two-sided sudden-death basketball - I think sudden death would be a good thing!). My hybrid idea, theoretically, allows each game to establish its own appropriate victory number based on pace of play and offensive skill.

Based on data, I do believe my proposed format would lead to more variance in actual game times than we see currently, but not as drastically as I initially feared. If a league is afraid to jump in to a new format with both feet (because they don’t want to upset the TV networks over varying game lengths), I’d love to at least see them scrap the five-minute overtime and adopt a first-to-ten-points overtime (or some other predetermined number of points). This eliminates the negative side effects of the game clock once overtime begins, and would actually reduce variance in game lengths among games that go beyond regulation (because there’s no possibility of additional overtimes).

Another format I had considered to derive the victory number for each game:

1. Teams play a timed, 20-minute first half (just like in the original post and in post #13)
2. At halftime, a flat, predetermined number (30, maybe) would be added to the leading team’s score to find the victory number for that game (unlike original post and post #13, where the teams’ halftime scores would be added together to find the victory number); if the halftime score is 17-16, play to 47; if the halftime score is 48-44, play to 78
3. Teams would play an untimed second half until one team’s score matches or exceeds the victory number

This format appeals in two ways (as compared to the format from the original post and post #13):
1. Simplicity - although adding two numbers (as in the originally proposed format) should seem pretty simple, this format doesn’t have any minimum score that would have to kick in during some games
2. (Perceived) greater motivation to score during the first half - this format would likely prevent some misguided coaches from regularly trying to keep first half scoring low (as they might do playing under the originally proposed format)

With that said, I still prefer the format from the original post and post #13 because IT appeals for many other reasons

I’m not sure this change would actually limit late-game fouling. What prevents the losing team from playing Hack-a-Shaq when Shaq’s team needs just a few more points to win (presuming a reasonably close game)? In fact, a winning team might do the same thing. If teams need to reach 96 and the score is, say, 92-90, the team with 92 is going to foul Shaq every possession, aren’t they?

When one team is a clear favorite and/or has a deeper bench, I would expect a severe clash of playing styles at the outset of the first half under this format.

It would generally behoove the favorite to boost the score and lengthen the game.  For example: come out in a full court press, push the pace, take every open three pointer, and perhaps even bait the opponent into taking semi-contested three pointers. 

Meanwhile, I would expect an underdog or a team with a thin-bench to try to use the shot clock and stay in a defensive shell.

If one team either builds a big first half lead or gets its key players in big foul trouble (with the score competitive), I would expect that team to slow the pace and emphasize defense.

---
In the second half, I would expect both teams to be more selective than normal in their shot-taking.

If one team has a multi-possession lead, I might expect to see them engage in basket hanging behavior.  If any defender(s) get caught behind the ball on a fast break, they can opt to turn around and make themselves available for a long pass to trade baskets.

The team that’s behind could play more zone and gamble on their opponents’ ability to make 3’s (no clock/tempo drawbacks).

As the magic score comes within reach, teams might adjust their emphasis on 3-point attempts vs. 2-point attempts vs. foul-drawing drives so as to reach the magic number as efficiently as possible.  Might seem slightly unfair to a team that is up, say, 89-88 with a magic score target of “90”.

---

It would be an interesting experiment.  But if the objective is purely to limit the late-game foul shooting parades, I think I prefer a less radical solution: Let the team that was fouled decline the free throw attempts and opt to take the ball out of bounds instead (and put a minimum # of seconds back on the shot clock if necessary).  We would still see more fouls than normal as the trailing team plays hyper-aggressive defense, but not so many long interruptions for free throws.

In response to post #17: Both the NCAA and NBA already have rules that make deliberate off-the-ball fouling (Hack-a-Shaq) pretty unappealing to the defense. I presume that’s why we see trailing defenses make such a concerted effort to chase down the ballhandler when they resort to fouling late in games.

NCAA: considers it an intentional foul (or a “flagrant 1 personal foul,” as the rulebook calls it) and grants two free throws to the fouled player and possession to the fouled team; this rule applies throughout the game

NBA: Grants one free throw to any player in the game for the fouled team and possession to the fouled team; this rule applies during the last two minutes of the fourth quarter or of any overtime (I think it should apply throughout the game)

With that said, I do foresee one situation where deliberately fouling the ballhandler late in a timeless format game would be more advisable than doing so at any other time in a timeless format game – if the defensive team is one or two points from victory and the offensive team is exactly three points from victory. I think the smart play is to commit a quick foul so as to prevent a possible immediate loss on a made three-pointer and to guarantee your team at least one more possession. In this situation, I would like to see the ensuing free throws granted technical-foul-style (any player in the game for the fouled team gets two shots with no one along the lane, but in this case the FOULING team would be granted possession afterward). This would avoid some goofy scene like the fouled team intentionally trying to miss free throws, etc. but that’s a discussion for another day. Anyway, it’s not perfect, but this 3/2-1 situation is more acceptable than what we see currently for a few reasons:

1.  It’s the LEADING team deliberately fouling; in a sense, they’ve earned the right to use fouling to their advantage (it still makes me cringe, though)
2.  This strategy wouldn’t be used repeatedly; as long as the fouled team makes at least one free throw, the game can’t enter this 3/2-1 scenario again
3.  This strategy would only be used in games where this particular score phenomenon arises; currently we see deliberate fouling in nearly every game that’s even remotely close and where the leading team has the ball late
4.  It would guarantee a thrilling finish – sudden death (or virtual sudden death) basketball

In response to post #18, the discussion of different styles (paragraphs 1-5, 7, 8) corresponds very closely to what we see throughout games currently. I think you’re right about a lot of them, but I don’t think there’s anything really shady about any of them, and it would be fun to see some of the different styles and strategies used based on certain strengths/weaknesses/situations.

I must admit I never considered the cherry-picking strategy. I might be overlooking something, but I don’t see how cherry-picking would be more effective under a timeless format than under the current format (after all, trading baskets has always been good for the leading team); or how it would be more effective than playing a more traditional way under any format (it seems like basketball players are able to get back to defend this pretty quickly; anyway, if cherry-picking does prove effective, maybe some type of defensive backcourt violation would need to be introduced)

As for allowing a fouled team to decline free throws and retain possession, this resembles some other late-game changes I’ve heard or considered in that it doesn’t give the leader any better option than stalling, doesn’t give the trailer any better option than fouling, and makes late comebacks from relatively slim deficits even more difficult than they are currently.

RedsGroundsCrew/20:

I don’t see how cherry-picking would be more effective under a timeless format than under the current format (after all, trading baskets has always been good for the leading team);

To illustrate by example: if you’re ahead by 10 with 5:00 to play, given a choice of a long pass that will probably lead to an easy layup, but carries some risk of an immediate turnover, generally teams will opt to melt 30 seconds off the clock and take an average FGA from their halfcourt offense while ensuring that they still remain up by 10 with ~4:30 left.  In a timeless environment, “melting the clock” would be meaningless and the only consideration would be whether the turnover risk + missed layup risk outweighs the average productivity of the halfcourt offense. 

----

As for allowing a fouled team to decline free throws and retain possession, this resembles some other late-game changes I’ve heard or considered in that it doesn’t give the leader any better option than stalling, doesn’t give the trailer any better option than fouling

Right. It would make basketball more like other timed sports like soccer or hockey or lacrosse: the leading team would try to play keep-away, and the trailing team would press the ball in desperation.  But the game would not degenerate into a slow ritual that is barely recognizable from the rest of the game. 

If the consensus is that 35 seconds is too long for a team to play keep-away, then we can tweak the shot clock threshold and/or revert to the old 5-second rule that was in place before there was a shot clock.

I don’t see clock-less basketball making comebacks especially less difficult as much as it would result in a different game flow during a comeback attempt.

(And this is a matter of taste, but I don’t prefer to make comebacks especially less difficult than they are currently.  I like that they are somewhat out of the ordinary and special.)

My first thought was to set the scoring limit at 100 points. If that’s too high due to some team variation then set it at 90 or whatever, and just have a different limit at lower levels. If we have different 3 point lines, half/quarter length, and field dimensions (in other sports) at different levels then we can have different final score limits too.

Also, if the ultimate concern is fouling at the end of games, then why not fix the foul problem instead of the clock? Trailing teams foul because it shortens the opponent teams’ possessions and hopefully also decreases their scoring.

The solution: Let the fouled team decide if it wants to take FTs or maintain possession with a reset shot clock (such as after a kicked ball). Trailing teams and both teams in close games will take the FTs just like they do now except in extreme edge cases (ie, fouled down 3 with 3 seconds left - 2 FTs aren’t enough). Teams with large leads will choose to keep possession and eliminate the trailing team’s incentive to foul. Teams with the lead will still want to milk the clock but that’s why there’s a shot clock.

Another solution is to let the fouled team decide who takes the FTs like technical fouls and soccer does with free kicks and penalty kicks. That eliminates Hack-a-Shaq and decreases the incentive on fouling at the end of games. My preferred solution is to let the fouled teams decide.

James/#22, I’m not sure your solution works. Let’s say Team A is winning and they never shoot the FTs, but always takes the possession and the reset shot clock. Team B would aggressively attempt to steal the ball, and if they commit a foul in the process, so be it… the clock stops anyway, and the opposing team doesn’t try to put more points on the board. Basically, they’ll just keep fouling (taking minimal time off the clock) while trying to either steal the ball or force a 5-second violation, taking at most 1-2 seconds off the game clock each time. I think it would actually lead to MORE fouls.

Tom N/23: The personal fouls would have to still accrue to the offending players, who would eventually foul out.  So that would be one limitation. 

Also I see it as a big improvement if:
(1) Team B commits fouls as a side effect of its desperate attempts to force a turnover, rather than as a primary strategy of fouling for its own sake.
And
(2) The pace of the game is not repeatedly stalled for 60+ seconds while FT’s are shot

James/22: Believe me – I would be all for completely clock-less games (when comparing it to many games under the current format), but I just don’t think TV would be too hot on the varying game lengths; and many blowout games might seem like an agonizing chore to finish, too (perhaps even more so than currently)

“if the ultimate concern is fouling at the end of games, then why not fix the foul problem instead of the clock?”
1.  Fouling is a big concern, but stalling is also a big concern. While the shot clock limits the extent to which a team can stall, stalling still makes up way too much of the offensive strategy for leading teams. Sometimes with as much as about five minutes left, stalling becomes almost as important as scoring when teams choose to burn a significant amount of time rather than run their offense right away. Within the last minute or so, running out the clock becomes THE focus of leading teams – they hardly even think about scoring. Not what Dr. James Naismith had in mind.

2.  Basketball certainly isn’t the only sport where stalling occurs, but it might be the only sport that could realistically abandon the clock that leads to stalling. When you consider almost any other time-based sport (field hockey, football, ice hockey, lacrosse, soccer, water polo, etc.), goals/scoring possessions happen so sporadically that they have to rely on a clock to keep game lengths within a reasonable range (imagine the possible range if World Cup soccer matches were played, say, first-to-two goals). Basketball was the same way 100+ years ago when only a handful of baskets were scored during a typical game. But for a loooong time, baskets have been scored much more rapidly - in the same way accomplishments are accumulated in other sports (tennis, volleyball, baseball/softball – outs as the accomplishment) that get by just fine without a clock. Basketball’s game clock might simply be outdated!

3.  Over the years, many people have proposed late-game rules changes to combat foulball/stallball that would still keep the game of basketball governed by a clock. I still question their effectiveness at truly encouraging teams to play by the spirit of the rules. But even if they are somewhat effective, they couldn’t produce some of the other appealing effects of timeless basketball: late-game clock malfunctions/errors/reviews (or silly endings like Tuesday night’s Michigan State/Wisconsin game) would be a thing of the past, blowout games would end more quickly and mercifully (I strongly believe), losing teams wouldn’t have to overtly concede games (currently we see teams eventually call off the dogs and stop fouling once all hope is lost – they actually give up before our eyes! I can’t think of this happening in any other sport, unless you count football when a trailing team chooses not to use their remaining timeouts while their opponent takes a knee), sudden death/virtual sudden death, and walk-off shots galore (currently, even many competitive big games in basketball lack a signature highlight; just think of some of the lasting images we’d get at the end of EVERY big game)!

I just re-read Devon/5: shortening the shot clock would be a very effective way to prevent dilly-dallying during the timed portion of a timeless format game. However, it might not increase scoring significantly (it’s been a while since I’ve looked it up, but I don’t think the early-90s change from a 45- to a 35-second shot clock increased scoring significantly, if at all)

But if first-half dilly-dallying were to emerge as the most popular strategy (I’m still not sold that it’s a sound strategy for most teams), shortening the shot clock might be a necessary measure.

Reds/25

1. If you’re also concerned about stalling, shorten the shot clock. That’s the exact purpose of the shot clock. I don’t see how you can expect to avoid end-game strategy in any sport - football teams run more, pass less, and score less when leading late. Soccer teams attack less and leave more on defense. Baseball brings in closers and subs in backups. When down to their last ‘out’, cricket teams only take even numbered scores so the same batter bats again. It happens in every sport, so if you have a built in way to increase offense then just utilize it.

3. Teams would still phone it in at the end of blowouts just like baseball teams put in their worst relievers and clear the bench. There would never be another overtime game, much less double and triple, because it is impossible for teams to tie. Also, shootouts would be a thing of the past - no more high scoring games 150-145 type scores between two uptempo teams pushing the ball. Good offensive games would end much faster and deprive fans of “full” games. Teams would slow down because pushing the tempo has no inherent advantage because games can end faster - you’ll lose the ability to wear out the other team down the final stretch because there may never be a final stretch! Meanwhile good, slow, plodding defense will always be an advantage because the game will last as long as you can stop your opponent. Not true for good offenses.

While from a competitive standpoint I like the idea of a clockless basketball game, and the attendant variable game length, would make standard basketball counting stats less uniform per game. Especially with the proposal in the original post, where teams may game playing tempo to try to increase their chances of winning. This isn’t necessarily a bad thing, but it would change how we think about headline stats.

Baseball is “clockless”, but it does have even time units: each team gets 27 (or 24, if they’re at home and leading ) outs. The outs are the “clock” of baseball.

Here’s another idea: why not count possessions, and play so each team gets, say, 150 possessions? The exact number could be determined from analyzing past data, and determining how long you want a game to last. That may be harder to do in practice, especially at lower levels, but for the NBA and NCAA division I it should be workable.

You probably could get rid of the traditional shot clock; there would be no more advantage in stalling on a possession, since the other team would get the ball eventually, and you’d have to make N stops.

This would put a premium on steals and defensive rebounding, plays which take possession from the other team.

It could be like match play in golf: in a blowout game, a team’s lead might be too much to make up. If you have a lead of more that 3 points per remaining possession, you’d win early, and not need to play the remaining possessions.

James/27:
1. Before there was a shot clock, situations would arise when wasting as much time as possible was at least as important as trying to score. Starting in the ‘80s, wasting up to 45 seconds was sometimes as important as scoring. Starting in the ‘90s, wasting up to 35 seconds was sometimes as important as scoring. If we shorten the shot clock, wasting up to, say, 30 seconds will sometimes be as important as scoring. None of these eliminates stallball. Basketball can do better.

Don’t get me wrong – as a fan, I love a lot of the end-game strategies we see in sports today. But I don’t love all of them. I couldn’t agree with you more about the strategies you mention: Football offenses spend most of the game trying to reach the end zone (that’s a good thing), but sometime during the fourth quarter, leading offenses run plays designed to burn clock more so than to advance down the field (I think that’s a bad thing; it’s less entertaining, but more importantly it ignores the most fundamental objective of the sport). The soccer strategy is similar in many ways. And late-game basketball might be the most excruciating of all to watch. Each of these sports is governed by a game clock, but basketball is the only one that could realistically abandon the game clock. As for the clock-less sport of baseball, teams spend the entire game trying to record outs and score runs. In the ninth inning, they try even harder to record outs (by bringing in a closer) and score runs (by replacing a batter with a superior pinch hitter) – I think that’s a good thing; it enhances the entertainment value, and it honors the basic objectives of the game. (I don’t know anything about cricket.)

3. In almost every sport, even when the game is seemingly out of reach, the players in the game for the trailing team try their hardest to help their team win (even that terrible reliever is doing all he can to get outs) and preserve some dignity. Well, there’s no dignity in playing foulball, and so basketball is currently one of the few sports (maybe the only sport) where the players in the game consciously avoid doing whatever gives their team the best chance (however slim) to win. I could be wrong, but under a timeless format, I honestly don’t think players would play halfheartedly and allow uncontested baskets.

And yep – so long, overtime! Basketball’s overtime doesn’t offer any new possibilities (it offers some more foulball and stallball, though!) – anything exciting that can happen in overtime could also happen during a regulation period. The overwhelming majority of basketball overtimes are anticlimactic (including the Heat’s 116-109 triple-overtime victory over the Hawks on Thursday – we waited all night for THAT?). And how could we expect any different? Any overtime period has an impossible act to follow (a period that was, quite literally, as evenly-matched as it could possibly be). Of course, even if basketball doesn’t make any OTHER changes, it could change its overtime to a points-based format – then it would offer foulball/stallball-free play, and even the possibility of sudden-death basketball. It wouldn’t be ideal (because I believe basketball should abandon the game clock for a significant part of regulation), but I think it would be a big improvement.

I think we would still see plenty of high-scoring games because I think teams have just as much incentive to score as they do currently (and for what it’s worth, even in the NBA it’s been 14 years since a team scored 150+ points in a regulation game). Under a timeless format, the points a team scores during the first half directly increases their opponent’s remaining workload. And that’s what scoring does currently – it raises the bar for the opponent. And you want the bar to be as high as possible!

As for uptempo and/or deep teams, they would definitely want to rack up points in the first half. Their best chance to wear the opponent down is to set up a second half that requires their opponent to score, say, forty-some points rather than twenty-some points.

I would strongly discourage teams from automatically trying to keep the score low in the first half of timeless format games – your opponent isn’t going to cooperate!

Geoff/28: This format has a lot of appeal - it’s another sound way to compel teams to play by the spirit of the rules.

Under this format, I would vote to keep the shot clock simply as a way to maintain a palatable pace of play.

I like that it also has a built-in mercy rule, and I agree that (3 * your remaining possessions) would be the appropriate deficit for it to kick in (even though a team can score 4 on a single possession when a player is fouled while making a three-pointer; or to be really technical, a team could score infinitely many points on a single possession if they are fouled, make the first free throw, miss the second free throw, offensive rebound, fouled, make, miss, offensive rebound…)

It could produce some awkward endings at times - this speaks to our mental math skills as a society, but there might be games where half the fans (and maybe some players, coaches, and officials!) don’t realize right away that a particular basket or defensive stop just put the game mathematically out of reach. Also, if the offensive team trails by three with only one of their possessions remaining, the defense would likely foul right away. This would force the offense to intentionally miss their second free throw to try to extend the possession. Or, at the start of such a possession, the offense may try to throw up a wild heave from the backcourt while being fouled to try to receive three free throws. Either scenario would be a little unsightly, and the strategy might start even earlier (offense down 6 with 2 of their possessions remaining; offense down 9 with 3 of their possessions remaining…) if the defense sees a chance to clinch victory. Maybe add a rule stating the game can’t end on a fouling possession if the offense trails by exactly (3 * their remaining possessions) at the start of their possession? And add on two more possessions (one for each team) after the free throws are granted? Not sure.

But those are just minor issues - the important thing is that this format compares very favorably with the current format in the way it addresses some major issues.

RGC/30 -

I had thought about the 4-point play, but rationalized that away by thinking that the other team simply doesn’t play defense any more, allowing uncontested shots.

And while I didn’t also consider your missed free throw argument before posting, to get to the line, you still have to be fouled first, so the same logic applies. The winning team simply has to stand around on the court and not foul anyone: they take possession after each made shot, so the “clock” ticks down even if you let the other team steal your inbound pass. Hence there is no need to play those extra possessions, just like we don’t play the bottom of the 9th when the home team is ahead, or the last X holes of match play golf.

You’re right about still needing the shot clock: while stalling for time doesn’t help you directly, if your team is in better aerobic shape, you might play “stall ball” to tire, or simply frustrate, the opposition, and become more effective later in the game. It might not be a common strategy, but surely some teams would try it.

Plus everyone is used to the shot clock now, so why remove it? While it’s not a strict cap on possession time, it does ensure things move along at a decent clip.

It’s funny you should mention overtime above, since it seems to me that in overtime teams are much more conscious of the game being a series of alternating possessions. Sure, it’s not a finite series, but it’s still a series.

Most of the drama of overtime isn’t from the deciding overtime period, but from the back-and-forth leading to that last possession to tie, or fail to break the tie, that results in yet another period.

While my original post didn’t specify it, you could also have an overtime of alternating possessions, just with a lot fewer of them, which should work out largely similarly to the current system. Except that you wouldn’t have to go through the last 2 minutes of garbage time if one team got out to an insurmountable lead…

RGC/30 -

Thinking a bit more about your fouling scenarios… one simple-to-implement, albeit highly radical, change would be this: when a foul results in free throws, let the fouled player keep shooting until a miss.

This would de facto mean certain players would almost never be intentionally fouled (and, in those rare instances when they were, it would be a rather long, boring stretch).

That’s probably too radical, though.

This reminds me of my year as an exchange student at a boarding school in Scotland. I was one of 3 American kids on the team, and in our first game the other team steals the ball and is going for a breakaway layup. One of the other American kids fouls him to prevent the easy basket, and I applaud him, saying “good foul”. My teammates (and coach!) look perplexed, as if we’re violating some honor code by intentionally fouling!

We are so used to fouling as a legitimate strategy, and indeed mandatory to prevent an uncontested layup, that we don’t give it a second thought, but from another perspective a “foul” is, well, just not cricket! So I’m not sure that you can remove incentives to foul in some cases so long as the penalties for fouling remain relatively mild.

But in the alternating possession game, fouling becomes a less attractive strategy for a trailing team overall, as you can’t squeeze out extra possessions by a lot of fouls, and increasing an opponent’s expected points from any one possession becomes an even worse tradeoff than it is now.

But I suppose this does reveal a weakness at lower levels of play: while an NBA or NCAA player is more likely to score from the line than from the floor, at amateur levels, especially early on, free throw shooting may be on average worse than field goal shooting in some cases. ...

Starting small might give clockless basketball the best chance of being widely accepted. Yesterday, I e-mailed a well-known innovative thinker within basketball (Mark Cuban) about the merits of changing the NBA’s overtime from a time-based format to a points-based format. His complete reply: “Its actually a great idea. We couldn’t do it in the nba but could be fun in the de league.”

Today, I e-mailed NBA D-League President Dan Reed about the idea (I also sent e-mails to the NCAA and NFHS) and I’m awaiting his reply.

I also remembered today that the American Basketball Association (minor professional league formed in 1999) currently uses a first-to-ten-points format when any game reaches double overtime. Clockless basketball CAN work!

I like that addition for overtime, and think it would make things a lot easier. There’d never be any double OT games and the like, but that doesn’t bother me.

Oh, having that rule in OT sounds like a great idea.

The NBA D-League is currently testing a three-minute overtime for possible use in the NBA as a way to avoid a dropoff in intensity from the end of the fourth quarter.

But this format will still be vulnerable to many/all of the drawbacks of the NBA’s current five-minute overtime (foulball; stallball; no guarantee of a signature highlight; clock malfunctions, errors, reviews; etc.) and will be MORE vulnerable to one specific drawback - additional overtimes will be even more likely. This shorter overtime may actually lead directly to longer overtimes!

A points-based overtime would prevent all of these problems.