record by people who don't have access to a dictionary.
Single-Season Records and Career Records
Most of these records appear the safest for the near term (Cy Young’s 511 career wins, Gretzky’s 92 goals in a season, Gretzky’s career points, Jordan's career ppg, Pete Roses’s 4,256 career hits, etc). The 4 major sports leagues have changed considerably since these records were set, and the current state of each league appears to not be conducive to anyone breaking these records. The decrease in the skill level gap between the best players and the worst has decreased as access to sports has increased, making the stars stand out less on the scoresheet than they did previously (or maybe we’re just waiting for the next great star). Would a 25 year old Wayne Gretzky be the best player in the NHL right now? Is Michael Jordan better than Kobe and LeBron? These questions are debatable, but would Gretzky or Jordan put up the same kind of numbers in today’s leagues? No. I'll let the naysayers know that Gretzky in his prime is struggling to stay on pace to put up 50 goals in my heavily simulated NHL12 season.
Even though the records appear unbreakable at the current time, structural changes happen in leagues that could blow-up the record books. The NHL could try to increase scoring through a rule change, or the NFL could make a rule that makes it much harder to defend against the pass. Potential future changes in medicine could bring about structural change as well: what if pitchers could pitch 130 pitches every other game? In addition to structural changes, league-wide trends can alter the context to break records. Dan Marino’s 48 TD passes in a season appeared untouchable for a long time; that’s not the case in the current pass-heavy NFL. Most importantly, the world has a while left in its lifetime, and a lot can happen in that time. Hopefully, we’ve only seen a small part of the major 4’s lifetimes, and their continued existence will allow ample opportunity to break all current records.
Single Game Records and Streaks
These are the most probabilistic, but often thought of as "the most unbreakable." DiMaggio’s 56 game hit streak gets a lot of attention. A .300 hitter has about a 76% chance of getting a hit in any game, assuming 4 at-bats per game. Assuming 4 at-bats per game, and ignoring the quality of the opposing pitcher, the probability of a 76% chance event happening 56 times in a row is roughly one out of 5 million. But that’s just any sequence of 56 games, so any player has (career number of games – 55) sets of 56 games. Even with 30 teams with 8 players (9 in the AL) trying to get a hit every game, this record could last a while, easily through our lifetimes.
The late Derrick Thomas holds the NFL record for 7 sacks in a game. If a top D-lineman has a 1/40 chance of recording a sack on a given pass play, and plays against 30 pass plays per game, the probability of any D-lineman with that probability of recording a sack who plays against that many pass plays per game is a little less than one out of 100,000. A player with a higher chance of recording a sack would be more likely to break the record, especially against a team that loves to throw the ball, but has a terrible offensive line.
The conclusion here is that these records are breakable in the current context of the leagues (although a player who hit .400 would be much more likely to break DiMaggio’s streak), but these records could last a while.
Longest and fastest
Unless the record can’t be broken (like a 100 yard rushing TD in the NFL), it will be broken. The longest distance play records are more a product of limited opportunities than being unlikely to occur-the probability of throwing a 100 yard TD pass could be around 1%, but teams rarely run a play from their goal line. Other records in this category are quickest goal off a face-off in the NHL, where these records could only be broken if measurement was in smaller denomination than seconds.
I think the more interesting “unbreakable” records are in the Track and Field, swimming, and fastest pitch category. These records are very breakable within our lifetime. Is Usain Bolt the fastest human ever? I think you can make argument that he is, based on the size of the current population and advances in nutrition and training. However, someone will break his record. There is no concern over the league structure of track and field and there is no threat from rule changes. With improvements in nutrition, Bolt could probably break his own record. The world population is growing, the percent of the population with access to proper nutrition and sports is increasing (well, I hope this is true), and sport technology continues to improve. Many track and field events, sprinting in particular, also benefit from an enormous pool of contenders because nearly everyone tries it.
The best part about claiming that records could be broken in the future? I can never be proven wrong, unless the world ends.