70 centimeters

This is a magic number that you may or may not know. 

70 centimeters and above for a katana is considered premium length. Below this is can be detrimental to the value of a blade. 

There are reasons for this.

Why should I care

The reason why 70 centimeters is a magic number is that this is fairly close to the average length for ideal use in the Edo period. We usually see koto blades cut down to between 69 and 70 cm. Sometimes a bit more, sometimes a bit less. Sometimes a lot less and sometimes a lot more but it gets more rare when you go in either direction.

The feeling is that these are lengths adjusted for use. Those blades shortened to 64 cm plus or minus a bit, I tend to think were Muromachi period shortenings to simulate the ideal length of current period katate-uchi, which were a katana meant for one handed use. These would be quite a suitable backup blade if you were using a yari on horseback. Not too bulky yet capable of fighting from horseback still up close better than a shorter wakizashi length.

When we get into shortenings below 60 cm, my impression is that these are done to correct damage to blades from fighting: cracks in the ha, major chips that are too much to polish out. So we can often see masterpiece blades by a smith like, say, Nagamitsu, that would have been 85 cm when made, cut down to 50 cm. With 35 cm taken out of the blade, it seems reasonable to expect that a crack in the lower part of the blade could be the cause of that, rather than just assuming someone wanted to destroy a great classic tachi and turn it into a wakizashi. Looking at the number of holes in a nakago when shortening may help explain why it was shortened. One hole = someone cut down a lot to get to that length, cutting out any prior used nakago in the process. Many holes stacked a few cm over each other is a machi slowly moving up over time, as people make length adjustments probably for comfort’s sake.

Anyway, if we measure the average length of unsigned suriage koto period blades that have passed Juyo Token, we get this number: 70.144 cm, which shockingly is right at the magic number we say is premium length for a katana when you buy it.

For us in the modern period, Edo period katana would be a little bit short to use. The users at that time were smaller stature than we are today, so we have a bit of a bias toward blades in the upper half of their ideal use range. And since blades cannot be made longer, only shorter, there is a natural pricing bias that comes with blades above 70 cm as they are perceived as more attractive and over time became more rare. 

This is data taken from about 4,000 koto swords that have been made suriage, and you can see the distribution does cluster around the 70 cm mark.

So, this one is advice that is not so hard to understand.

When you pick a sword that is over 70 cm, you’re getting one that’s on the larger size than average. As you add 5 or 6 cm to that, it looks like you quickly end up in the top percentile when it comes to a suriage mumei koto sword.

Why, do you ask, am I looking at shortened swords to get an idea of what is premium length, instead of unaltered swords?

There are a couple of reasons for this.

I didn’t want to look at koto tachi, because they are either original length swords made for earlier eras and preserved unaltered as artifacts of those areas — or, they were altered not necessarily for ideal use, but in some cases just short enough to not damage the signature. So the owner in this case may have ended up with a shortened sword he found a bit awkwardly large, but he made sure that he retained the record of the maker on the blade in exchange for that. This isn’t giving me information about what length he prefered to use, only the minimum length required for that tachi to still retain its signature on it. As for Muromachi katana, the problem is the opposite, in that many were made a bit small and they may tilt the scales in the wrong direction then when counting them.

I am trying with these numbers to get a handle on what the average Edo period samurai would find to be an ideal length for a katana, so I looked at koto blades “custom fit” for Edo period use.

Well then, why not look directly at Shinto swords if we want to know what an Edo period samurai wanted to use? 

The answer for that one is…

Selection bias

In order to get a handle on sword lengths it’s necessary to get a pool of data to look at. I don’t have all the Tokubetsu Hozon or Hozon ranked Shinto blades to access but I can check the length of all the Juyo shinto blades. 

Given that we already have a bias toward longer blades, and Shinto swords have “more to prove” in terms of condition, beauty and quality than Koto blades, which also have age going for them, I theorized that length will be a bias that comes to play with Shinto swords. 

If this theory is true then I would expect the average length of Shinto swords to be longer than Koto unsigned suriage swords cut for comfort and this does bear out in the data. The average length of a Juyo Shinto sword is 72 cm, quite a bit longer than a koto suriage blade.

So, I think this lines up with my theory that if you’re bigger than average it is going to help you pass Juyo as a Shinto blade. A Koto blade the length is good but since “what we have is what we have” after all of the shortenings, length becomes a bonus but not everything for passing Juyo as koto.

This is the distribution chart for Shinto blades. 

Interestingly, the distribution of lengths is not the same shape as it is for koto blades. There are two clear humps here showing that there is a secondary preferred length centering around 75 cm.

Drilling down and examining these blades between 73 and 77 cm they seem to come mostly from the Momoyama period when copying Nanbokucho blades was in vogue, and once again in the late 1600s, for reasons I can’t yet guess at. This is a general tendency and the data might be getting a bit thin to make that assumption but the chart seems to clearly show two ideal clusters of length, centered around 70 and 75 cm for blades that are passing Juyo. So, there seems to be some need for blades made in the Shinto period to have this 75 cm length that was lost when most of the long koto blades got cut down. 

My feeling is that those 75 cm Shinto blades remain unusual, and if more standard Shinto blades were added to this, blades that were utilitarian and didn’t match Juyo characteristics, we would continue to see that spike around 70 cm grow higher while the 75 cm spike gets drowned out. I think there is a take home here that this type of sword is unusual and special, and is assisting the blade in passing Juyo.

For their own part, Shinshinto blades continue with increasing average lengths in their presence in Juyo, averaging 73 cm. I didn’t bother charting them because there are not so many Juyo Shinshinto swords (only 356 katana) and I’m unsure if this is enough blades to make assumptions, other than that it seems again that longer sizes are helping select those blades at Juyo. 


When it comes to a koto blade, lengths above 70 cm are indeed premium and go on the list of nice to have attributes of a blade. Between 68 and 70 cm the length is still good and not so much anything to be worried about. 66 to 68 cm blades are somewhat short and the length starts to take away from the value in the market as we are starting to creep down the steep edge of the curve. Once you go below 66 cm you are into the short zone and there is a definite impact to the value of the blade. 

With Shinto katana, there are always more choices available and so the ideal length applies more strongly. When you start going below 70 cm it is easy to wait for another option, and many buyers do. This impacts these blades. This is not necessarily a good or a bad thing, it’s just something to be aware of.

I am not saying a 70 cm blade will always be more valuable than a 69 cm blade. It is just one dimension, an objective dimension, that we can use to compare two blades. When lengths are close, the quality and preservation of the blades, if there is any difference, will easily trump the difference in length. But when the length difference is extreme, it becomes a large factor and when supply of available examples is a non-issue, say for a smith like Omi Daijo Tadahiro where there are many thousands of examples out there, then a katana being short causes a major strike against it in the marketplace.

This can help you because the price should fall considerably to make up for this detrimental aspect of the blade: you can buy then a blade and save a few thousand dollars in exchange for a couple of centimeters in length. 

When you are considering a smith like Masamune however, you do not have a lot of options available to you. The NBTHK has only authorized Masamune in katana form at Juyo 30 times. 

Of those 30, about 25% have some doubts written into the commentary, indicating that they are work of one of the highest skill Soshu smiths but Masamune is a conclusion that is only tentative and maybe not the best conclusion. The majority though the commentary indicates firm acceptance of the attribution. 

Regardless, if you are in the market for a Masamune, and one comes available to you at a price you can afford and the length is 67 cm, then it is not something you should be looking at or even worrying about. You may only have one opportunity in your life to buy such a blade, if you can afford it. The same reasoning goes for any smith or school who has very rare work. 

For example, one of the finest blades I ever saw is a Tokubetsu Juyo Awataguchi blade that is cut down to 65 cm. The preservation in the jihada is perfect and the quality is absolute top grade for any school or maker. To turn this blade down on the basis of length is foolish, because there are almost no middle Kamakura blades that exist with this level of preservation at all. So it would possibly mean discarding this blade in favor of another at 70 cm that was a lot less healthy… you’d get a longer blade but lose the health and that wouldn’t be a good choice. If however you found one that was 70 cm and equally healthy then this longer blade will indeed be more attractive (and also more expensive) than the 65 cm one. 

As such, length absolutely is not everything. It is only one objective property of a sword that factors into value. You must always consider it, you can use it to differentiate blades that are otherwise identical. 

You have to also make sure that you are not lead around only in a quest for the biggest of swords… as you will surely miss out on great masterpieces if you do.