You don't need stronger muscles, you just need fear

The difference in strength between the strongest and average human is really not that important in the scheme of things. Especially in sword fighting, even the most compact muscles can easily be punctured by a blade. In a room full of swordsmen, a lucky strike can shear through to bone, and then the master swordsman is done.

On the other hand, remember that scene from Unforgiven? You might say it is patently impossible for one man to kill 7 in a shootout without being hit back. But that is explicitly discussed in the movie. Kill the leader first, keep up the attack, rely on fear and a bit of luck.

Using swords and wearing armor makes this even more possible. Strikes from a fearful arm lead to glancing blows. The intimidation factor of the man next to you being shot is nothing compared to the man next to you erupting in a fountain of blood as someone cuts his torso nearly in half, etc.

A plausible no-magic scenario

In one corner we have “the hero”. He is truly excellent with his sword, though not immortal. What he does have is practice dealing with several opponents at once.

In the other corner we have six bandits. They are by no means unskilled, even if each one would lose a fair fight one-on-one with the hero. They, on the other hand, have not trained how to fight six against one effectively. (Why should they. Who loses six-against one? In such situations, the opponent just gives up, pledging for mercy right?)

The hero is lucky in one aspect. The ground is flat. As long as his back is clear he is free to move about. He will not become cornered.

The fighting starts. Six bandits rush against the hero. The hero does not fight all six. He backs and steps left getting out of the way of most of them, parries, ripostes and cuts one bandit badly. Another swing while the bandits are confused, six opponents has become four and the hero backs away quickly.

By now the bandits have realized that the opponent is dangerous. They try to spread out, but since they are not coordinated, the hero manages to move about so that he never faces more than one at a time. One bandit hesitates, the hero attacks and it’s three to one.

This is where mind game takes over. A determined, coordinated attack by three would still mean the end of our hero, but while all three bandits are willing to move around the hero to cut him down from behind, none of them wants to be the one keeping him occupied head on.
Maybe they run away, maybe this indetermination causes the end of one more.
Fight is over.

Why did the hero win?
He was prepared, bandits were not. If two men had attacked him head on at the beginning while the rest flanked him from both sides, he would be dead meat.
Also he had no choice but to fight full out to survive while each bandit could hope that one of the others took the lead (and risk).

Would it work every time? Hell no. But a hero is allowed to be a bit lucky.

Looks like I'm late to the party but – hello! I'm a former nationally ranked saber fencer and a coach of many years.

It turns out that it's actually *easier* to fight many swordspeople at once than individually.

In my (anecdotal) experience, there's nothing unrealistic about your scenario! If there's even a mild difference in the skill levels between the one and the many, this is very reasonable.

When I was teaching at the college club level, after my students had had a few hard days of training, I'd add in a game day for them. Nothing like how fencing actually works in competition, just fun activities to make them feel like swashbucklers. Most often it would take the form of dividing them into teams and letting them fight it out melee-style, but occasionally they would convince me to do a many-on-one type game with me against all of my students. These classes could be up to a dozen, dozen and a half students, so it wouldn't be out of the ordinary for me to be taking on 15 people.

I won almost every time. I'd like to take credit for that, but in fact, very little of it had to do with the raw skill differential.

In a one vs many engagement, if the one keeps moving, the many will end up getting in each other's way. The one is unconstrained in their movement, while someone in the group will have less freedom of movement, on average, because of all the other people pressing around them. As any swordsperson will tell you, if you're not able to move around freely, your opponent will get the better of you 9 times out of 10.

The key is this: the group interferes with itself, so even above average fencers will be at a disadvantage against a single competent opponent.

I know it seems counterintuitive, but I see it over and over again. You don't have to take my word for it, though. Here's video from a while ago that might be informative, wherein 3 olympic fencers take on 50 opponents at once:

https://youtu.be/PgKg0Hc7YIA

Yes, it's a contrived example, but the principles remain the same. You'll note, the experienced fencers don't start getting knocked out until they're fencing more or less one on one.

Let's consider something more reasonable.

A heavily armoured knight taking on half a dozen ordinary infantrymen

He's probably got a decent chance here, he's a professional soldier, an ordinary infantryman is a peasant with a pointy stick. This is no meeting of equals, it's someone whose sole purpose in life is to train to kill people better versus a group of farm boys who wanted to see a bit of the world before going home and that's if they were volunteers rather than conscripts.

Of course the telling of the tale in the bar afterwards, there were more of them, a couple of drinks down the line and they were bigger, by the end of the night they were also elite knights.

And so legends are born.

As good as he may be, no one man is going to be taking on a room full of "elite" swordsmen. His blade can only be in one place at a time and their blades will be in many places. Even if he can defend against them all, he will fall from exhaustion before he can despatch them.

There are some good answers here, especially the fear factor. However one aspect that has not yet been touched on is technology.

However you cut it, one swordsman against many is a highly risky proposition that you wouldn't expect to work more than once or twice. But give that swordsman greatly superior equipment, and it becomes more likely. Imagine being a bronze-age skirmisher coming up against an armoured knight wielding a steel blade: your spear/dagger can barely hold an edge, while he's cutting through your cohorts like butter and every blow they manage to land seems to bounce right off. Even a group of expert warriors would think twice.

All else being equal the equipment of a true professional knight gives them an enormous advantage over rank and file soldiers who were the medieval equivalent of modern territorial soldiers; farmers pulled from the plow, given minimal training and second rate gear if they were lucky. Full plate armoured knights were seriously hard to damage even for fellow fully armed and armoured opponents.

Added to that you have training; the real masters trained to a very specific degree, as much as they could without damaging themselves, from as early as they could without damaging themselves, as continuously as they could without damaging themselves, for as many years as they could keep it up without damaging themselves.

On top of that you will always have a few statistical freaks who can out perform most of their peers at a given task.

If you combine the effects of exceptional natural talent with highly effective training that starts at 5 or 6 years of age you get a 20 something year old knight who, while in their prime, can take on anything you throw at them. Add to that the money to get the best custom equipment and they're more than a match for other knights, and a lawn mower when it comes to conscripted farmer-soldiers. There won't be many of these people around in any given place at any given time (maybe 1 in 100 knights perhaps 1 in 10,000, maybe less, combatants in a major battle) but you would not want to cross blades with one of few who were.

Sorry to answer your actual question:

• yes if you increase maximum performance for strength, stamina (very important in battles less so in tournament matches), and reaction times for the very best combatants you will get a sharper, scarier, peak when it comes to best performance and slightly increase the incidence of "above average" fighters. Making the absolute best fighters swift death to everyone else and creating a larger elite pool for them to be measured against. So instead of the "best of the best" being 1 in 100 knight they'll be 1 in 10,000 with another 500-1000 in ever 10,000 knights who are the "best of the rest" being as good as the very best would otherwise be.

It is indeed realistic, mainly because most humans fear death, and it probably happened plenty of times in history.

We can't really take sport fencing or computer game mechanics as a basis. If they were fighting with training gear fully protected, or if they were mindless drones where the master swordsman differed only in physical strength and slightly better reflexes, then yes, even just 2 or 3 lesser skilled opponents would have a great chance of defeating a single more skilled opponent.

First, let's assume the fight doesn't start with the master surrounded, because that would reduce his chances enormously. One of the most important factors would be how the fight starts. Most people don't realize instantly that the situation is a life-or-death one, and they will lose precious seconds in the confusion.

1. Most people lack the mindset and the intent to kill

There was a research shortly after World War 2 which showed that a huge percentage of soldiers were not even firing into the direction of the enemy, and even those who fired, very few of them were aiming at individual soldiers, most were just firing wildly in the general direction of the enemy. If the opponents are just soldiers pressed into service, or they are bandits fighting for loot instead of a cause which would motivate them to kill and to risk being killed, it would look like most bar brawls are: they are more into scaring their opponents away or dominating them, than outright killing them.

2. Most people fear death

Those of you who ever sparred, practiced sport fencing, or just LARP, would you go into the fight with the same eagerness if you knew it was to the death? Probably all of you were hit at least once during practice. Imagine it was deadly.

Technically, if several less skilled swordsmen lounged forwards at the exact same time, one would be surely killed by the lone master, but the others would manage to kill him in return. But are they coordinated that well? Probably not. And who is the one who wants to sacrifice his life to go first? Especially if the attackers are bandits, who don't fight to protect their loved ones, so they don't have anything to sacrifice their lives for...

Probably, as the fight starts, everyone expects someone else to strike first, and a precious few seconds will pass during this time of indecision. There can be at most 4 or 5 in the front facing the master directly (any more and they would just get in each other's way), the rest would be behind them. Now as they wait for someone else to strike first, the master swordsman quickly finds the one or two whose guard is the most vulnerable, and stabs them before they can react. They don't fall to the ground like sacks of potatoes, like you can see in movies or video games. They will writhe violently on the ground, screaming in agony at the top of their lungs, with blood spattering around. Some would vomit at that sight, some would run away, and many of the rest would be frozen for a short moment, allowing the master swordsman to strike again.
At this point those who are still uninjured, would probably fear for their lives and just wildly flail around with their weapons, not really eager to close the distance. They will be uncoordinated, allowing the master swordsman to engage them one or two at a time.

4. Fight-or-flight reflex.

It takes a lot of training to overcome basic instinct. If the fighters are just averagely trained, in a life or death situation it can happen that all their training and all their techniques go completely out, and they are back at basic animal instincts.

5. Armor.

This is what makes a great difference, especially when combined with the fight-or-flight reflex mentioned above. Fighting against armor with a sword requires very elaborate techniques and great precision to stab at the more vulnerable spots of the armor, and there are complex techniques for getting into a position to be able to perform those movements. You can't defeat metal armor by just wildly hacking at it with a sword. If the fighters revert to basic instinct, they will not be doing much damage against someone who wears armor. If the master swordsman is really well trained, his mind will not be affected, and he will still be able to perform the techniques needed to defeat his opponents even if they wear the same quality of armor.

6. Coordination

Even if the attackers are individually not too bad at fighting, they might lack the skills and the practice of fighting in a coordinated way, and they might just get in each other's way.

Here is a great video demonstrating many of the issues mentioned above: https://www.youtube.com/watch?v=7s7KfetTixQ

The main problem is not reaction time nor the time to grow muscle.

The main problem is that swinging a sword around is an heavy job, and the stamina one has is limited. It doesn't take long before one side is puffing and panting, and that usually mean death for that side.

In a one to many combat scenario, the one can at best take down one, maybe two opponents, but then will be simply left out of fuel, and spitting a watermelon seed on his forehead will take him down (well, not literally).

I think the most realistic way to go in the direction you want to go is to improve stamina and endurance. Kind of a marathon training. And still it might be better to use that training for fleeing and not for fighting.

Armor and training is a big factor.

Even in your Selmy example, he actually had an entire scene of removing his armor, and still had more on then most of the room. With enough training, practice, and armor, yes a single knight can take on an entire room. It's happened frequently in history (reference the Templar knights).

Stamina is a big factor, but so is technique. I once went to an archery range and this old guy in his 80s was tending the counter. I asked him about this bow, and I couldn't even pull it. He told me it was set to 90 pounds, he then shuffled up to me, weakly picked up the bow, then somehow smoothly pulled it back and fired a shot at the target. That's an example of skill trumping strength. He knew the exact way to position his muscles, whatever remained of them, to pull that huge bow, hold, aim and fire it with little trouble. I imagine a master swordsmen is the same, they would know exactly how to swing and parry with maximum effect and minimum energy expenditure.

Also having the correct weapon. In fantasy, everyone is swinging with their mounted swords, that's unrealistic and would be very tiring. But a musketeer would easily dispatch Selmy because their weapon is meant for getting between the chinks of the armor, and their own armor is designed to take arrows and bullets and swords. Despite how unrealistic the show was, Deadliest warrior actually did show this concept very well with the spartan battle and having equipment designed to kill with every strike (again not like Hollywood). Even their shield is a deadly weapon.

There are some real historical techniques to deal with multiple enemies, you can look up the "Montante" or "Spadone" sword school in historical european martial arts (HEMA). The Montante is a way larger than normal sword used in the iberic countries that is simmilar to "Zweihanders" or "Claymores", that were used in late medieval and renassance times. Their techniques involve dealing with a higher amount of enemies and even protecting others while doing that.

About the fear factor that was pointed in other answers: there are accounts of giant sword users like Pier Gerlofs Donia for exemple, that just by his own absurd size and sword size could make men shit their pants in battle, and with proper techniques you don't even need to be big to use the weapons.

Video from a guy who trains this kind of sword techniques:
https://www.youtube.com/watch?v=vxHaNRO705k

I'm a trained Iaidoka (samurai katana ritualized fighting) as well as a Live Action Roleplayer, where medieval-style combat is common.

A trained fighter can absolutely take on a 3:1, 4:1 or 5:1 group of enemies. In fact, in Iaido there are many katas that assume you are facing multiple opponents and how to cut down three or four enemies in about as many strikes and half as many seconds.

In LARP fighting, we once held a narrow pathway with four people against an entire enemy "army" (about 50 real people, but with respawns).

Location and strategy are vastly more important than muscle or reaction speed. A narrow passage can be held by few against many, and sometimes even by one. There is the legend of Horatius Cocles, who is said to have defended a bridge into Rome alone against an entire army (at least for a short time).

The trick to every successful battle of one-against-many is to fight sequential one-on-one battles, i.e. you never actually fight all of them at once.

So, to sum up the other excellent answers, and to address the OP question:

Trainable reaction time and more efficient muscles would have little effect in a one vs many battle. A blade and field of vision can only cover so much flank. BUT the psychological aspect of seeing such a proficient fighter with a high reputation would have an impact on the willingness of a member of the group to engage in a battle with said legendary hero. Further, tactics and situational awareness combined with these physiological improvements would allow a swordsman to last longer in one on one fights and if he could prevent himself from being surrounded could take out more enemies than he would have otherwise.

Strategy is King

If you just make your fighter and faster, all you do is tweak the number of people they can take on. You can always just pile on more people. It takes strategy to win against an unbounded number. Physical acumen can help, but strategy is key.

Hence the saying: If you know the enemy and know yourself, you need not fear the result of a hundred battles. If you know yourself but not the enemy, for every victory gained you will also suffer a defeat.

If you know neither the enemy nor yourself, you will succumb in every battle.

• Sun Tzu, Art of War (chapter 3)

The master swordsman is not just a swordsman. He is an illusionist as well. He knows that the fight is not just in the flesh and steel, but in the minds of the fighters as well. Through knowing himself, he has learned to know others. And thus will not fear defeat in a hundred battles.

The master swordsman is a survivor. He does not enter a battle unless he already knows he can win. This means he has already seen the loss in the minds of the other fighters.

We know this because we see "taunts." We see attempts at intimidation. These are approaches which try to make the opponent think they have already lost. However, the master swordsman approach must be more subtle. It must find the loss that already exists in the opponent's mind, and strengthen it until the opponent cannot ignore that illusion.

IF they do, the enemy hesitates. If they do, the swordsman can pair enemies off against each other. Indeed, the typical approach to winning such fights (in the movies, of course) is to use the bodies of one's opponents to get in each other's way while the master swordsman moves freely.

This kind of skill is not easily gathered. Indeed we typically think of it as impossible, which is why it is relegated to the awesome fight scenes in movies rather than the history books. But it is technically possible. If you can get in your enemy's mind and strengthen the false preconceptions they might have about the world in your favor, it really doesn't matter how much steel they bring to the fight.

Of course, it's easier with Magic (or just being awesome like the Doctor). I wrote an example of what it might look like in an answer about Alynn the mage who sought to try to hide her telepathic and telekenetic abilities so that she wouldn't be run out of town. I can't say whether it is good prose or not, but I had fun writing it, and it does try to capture what a fight might look like when an opponent's mind is already lost.

I want to point out a few things that might help add to the answers, but may not be important enough to be a primary answer.

Anticipation

If you want a character who is able to best several elite swordsmen, then they should be able to anticipate what elite swordsmen will do. This character should have meta knowledge of how elite swordsmen fight, and should be able to anticipate the ways in which they always consistently strike, and instead of reacting to their movements after the movements are seen, would counter the movement that he knows they are going to make. If there is variance in the options that an elite swordsman might have (i.e. if the next move is unpredictable), then this master would avoid the unpredictable situation entirely by taking a step back and moving out of range, then moving back into range in a way that creates a more predictable state. This master should have spent extensive time planning out how to counter standard attacks. He should be very perceptive to slight tells in the opponent's next move, such as the initial contractions of muscles and the dilation of pupils. This master should also set traps, e.g. by feigning an exploitable weakness, and then countering the expected exploitation of his feigned weakness. The setting of traps is something that should be well-rehearsed.

Group Tactics

The master should also use tactics that attack a group of opponents as a whole, rather than as numerous individuals. He should use multi-purpose maneuvers e.g. parry one sword in a way that forces it to stab a third party or in a way that forces it to parry a third party's attack. If he is not well-positioned for multi-purpose maneuvers, he should make slight lateral movements to a position that is more ideal for a multi-purpose movement. The master should have thoroughly practiced multi-purpose maneuvers, and should have multiple training partners willing to help with this practice. The master should make use of space and footwork so that the adversaries are bumping into each other or failing to cram into narrow spaces together. The master should make use of movement to corral the opponents into a group and fight the front opponent in that group, rather than being surrounded by them. The master should never be surrounded in the first place, because as soon as people start to surround him, he should sprint towards an exit point to get outside of the circle; and as an uber master, he should have a sixth sense for when people are beginning to surround him. He should run through doorways and then fight the opponents as they attempt to filter through the doorway. He should utilize numerous non-sword attacks and tactics such as slicing chandelier tie-downs, flipping tables (flipping a table on its side can create a barrier for movement that can prevent opponents from efficiently circling you), and splitting inflamed alcohol.

Visualization

If you are surrounded by numerous elite swordsmen and having to defend strikes that are coming from behind your back, then how can you approach that? Eyes in the back of your head? This master should have advanced visual memory that can keep track of the location of opponents who are standing behind him combined with the ability to anticipate how they will act while they are behind him. This should be honed by way of difficult visual exercises as well as practice with swordfighting a training partner who is standing behind him.

Stamina

To make a parallel to boxing, boxers who are experienced at the world class level (even if they may be experienced in losing at that level) and experienced at going many rounds can potentially "take an opponent into deep water and drown them." What this means is to use stalling and damage mitigation tactics to force the fight into late round (e.g. beyond round 7) and then beating them with superior stamina and having paced themselves better. In a similar vein, I propose that the master swordsman use stalling tactics such as defense, running around, and generally evading to draw the fight out longer. In a very long fight, you can leverage your stamina much better. Also, making a parallel to certain strategy games with luck components (e.g. Magic: The Gathering), a superior player's skill advantage can sometimes be trumped by luck, especially in a short game where few decisions are made. But if the superior player plays defensively and draws the game out into a very long game in which hundreds or thousands of decisions are made, the less skilled player will inevitably make dozens or hundreds of mistakes, and the more skilled player will win. So I propose that if the master swordsmen uses stalling tactics and movement to produce a very long fight, he is also likely to be presented with a vast multitude of opportunities to capitalize on mistakes. This applies especially when fighting larger numbers of less skilled opponents (yes, sometimes he needs to defeat 8 elite swordsmen, but aren't there also times where he will have to defeat 39 intermediate swordsmen?). It should not be assumed that most medieval elite swordsmen are also world class endurance athletes. Therefore, there is the opportunity to gain an advantage if the master IS a world class endurance athlete. My opinion: he should sprint one mile before beginning 6 hours of sword fighting practice, and jog another 7 miles after practice.

Your main problem is that swords are reasonably deadly and there's a limit to the movements a human body can perform. One grandmaster will almost certainly lose against a room of elite soldiers if he isn't wearing full plate armour or can use a chokepoint to fight them one at a time. That is because they only have to hit him once in order to win and even if we assume that he won't ever make a mistake or get exhausted, they can surround him and attack multiple places at once to make it impossible for him to parry or evade. Being able to see the attack coming won't help him there, nor will being strong or moving fast unless he is several times faster than an "elite".

If he's wearing full plate he might have more of a chance, but it's still possible for them to pile on him and pull him off his feet, at which point he is doomed. Assuming they're not wearing any armour though, he might be able to cut them down while positioning himself cleverly enough to remain standing.

A combatant who is drastically stronger than the opponents they are facing has access to a huge advantage: reach. The degree of strength that you would need would be bordering on supernatural, but wielding a very large, long weapon while maintaining elite skill and speed with it might enable a master to take on multiple opponents successfully, especially if they can do it 1 handed (2nd weapon or shield in the other). And keep in mind that reach is a function of the size of the fighter, so a large weapon plus overly long arms can give a huge reach advantage over an average person. Look at some of the wingspan measurements of NBA players to see just how extreme the proportionate length of a person's arms can get.

And in line with what others have said about the psychological aspects, facing an opponent with this kind of reach advantage would be terrifying. Trying to close the distance with them as they moved away from you, attacking you freely while you can do nothing but defend yourself. Who is going to throw themselves at that? Even a surrounded fighter with this kind of advantage might be able to create enough room around them to attack and breach the encirclement and manouver for better position.

This seems like a way to create an "untouchable" master, even a large group of opponents would almost have to throw themselves into suicide knowing that a number of them would certainly be killed in an all-in rush. What is their motivation to do that?

Allow me to direct your attention to Cyrano Bergerac. Yes, that Cyrano Bergerac. Yes he was a real person. Yes, he really did fight duels over insults to his nose (though not as many as the play would have you believe.) Yes, he really was a master swordsman with several really nasty fencing tricks credited as his inventions.

More relevant to this question: as far as we can tell, yes he really did once take on an entire gang of armed men all by himself. One of his friends had offended a nobleman, and the nobleman had hired 100 men to stand at the town gate and wait for the poor soul to try to go home so they could murder him. (The number may be exaggerated, but probably not by that much. The goal was to show off how powerful the noble was to be able to afford such a lavish way of disposing of a single nobody.)

Cyrano would have none of this and escorted his friend home. According to the witnesses, the result of the battle was two dead, seven wounded, and a great many pikes, sticks, and hats discarded by the rest as they were attempting to flee.

https://archive.org/details/voyagetomoon00cyra (The translator's note contains the story in English with the names of the original reporters if you care to investigate further. Also the book's not half bad if you like Renaissance Sci-Fi.)

So yes, a single, well trained psychopath with the element of surprise, little fear of death, and a demonic grin on his hideous face can certainly overcome an entire room full of fighters. The key is to get that panic started quickly. Once it takes hold the opposition will dry up and blow away like leaves. The difference between a fencing expert and a fencing master is that the expert fights their opponent's body while the master fights their opponent's mind.

When Legolas or Captain America or whoever is surrounded by a half-dozen men, he doesn't keep parrying them all until he can take them down one by one. Instead, he'll spot an opening to, say, jump off the outstretched leg or sword of one attacker and vault out of the circle to catch his attackers from behind.

Of course the movies do this because it's more dramatic, and also easier to demonstrate visually.

But as a side effect, it often feels less implausible, not more, than the traditional alternative. One man parrying six blows aimed at different parts of his body from all sides in the span of a couple seconds? Ridiculous. Having the superhuman reflexes to spot and take advantage of an opening that a normal man couldn't? That's easier to accept.

(Of course when poorly done, it just looks ridiculous. But just don't do it poorly…)

I don't think the reflexes need to be at the superhuman level of an elf or steroid super-soldier or whatever. Just being a little outside the realm of real-life human experience is enough, because we don't have solid intuitions of what's possible outside that realm, so we can suspend disbelief more easily.

Especially since being even a little bit superhuman is going to have the same effect in-universe. If you just saw someone do something that nobody you've ever sparred with could do, you have no idea what else he can do. All the tactics you know could be completely useless. A smart master swordsman will exploit that. So he won't have to fight 100 average swordsman, he'll fight a couple of average swordsmen, then a couple dozen worried and tentative swordsmen who are much easier to take, and the rest will just run away.

I'm not sure more strength will work for that, because it just shifts the whole scale of expectations that your world's swordsmen grew up in. But making reaction time unboundedly trainable? I think that could be enough.