Daily & Intraday Maximum Drawdown

[postscript]

New to the forum, my first topic, so please be gentle!

I recently got interested in the Permanent Portfolio having read Fail-Safe Investing and then Rowland & Lawson's book. Was pleasantly surprised to find this forum dedicated to it, and looking forward to participate more in the future.

I've searched the forum for information about the topic of drawdown, but came up somewhat blank, so thought I'd create my own.

Does anyone know where I can get any sort of information on drawdown (on end of day basis - ideally by asset type) for the Permanent Portfolio, and even better on an intraday basis (although I realise that's a long shot). I really want to understand the maximum drawdown.

I have an account with AllocateSmartly, and based on the way they calculate it, I know the worst end of month drawdownwas -13% was 03/1982 and worst monthly return was -8.7% in March 1980, and the second worst was, more recently, -8.4% in Oct 2008. Other places online have calculated worst drawdown as nearer 18% and probably these were based on an end of day basis. Although I wonder if intraday, the loss could have been even greater, say 20% or more.

I'm aware in an ideal world, we'd only be checking our portfolios once a year, but it would give me piece of mind to know what the very worst historical drawdown has been. If anyone can give me some pointers, it'd be much appreciated.

[stuper1]

There was a guy who used to post on here a lot who would claim that the maximum historical daily drawdown was 25% if I recall correctly. People challenged him on where he got that number, and to my mind he never answered very convincingly.

Anyway, it's good to know that such things can happen in the short-term, so that if they do happen, hopefully I will just be patient and let things ride, and in a few weeks they probably won't look so bad. The losses only become real if I sell at a low point and lock them in.

[barrett]

Welcome, postscript! Just checking in on my phone and can't run the numbers right now. Go to peaktotrough.com and look at 1980-1982. Make sure you have the settings the way you want them and run the daily numbers. THEN, pull up monthly inflation numbers from that time period and you can get a pretty good idea of the largest total drawdown in real terms. IIRC, it was something like a nominal drawdown of 15% with another 10% hit due to inflation. And that was over an 18-month period. Please let us know what you come up with.

[Xan]

There was a guy who used to post on here a lot who would claim that the maximum historical daily drawdown was 25% if I recall correctly. People challenged him on where he got that number, and to my mind he never answered very convincingly.

Anyway, it's good to know that such things can happen in the short-term, so that if they do happen, hopefully I will just be patient and let things ride, and in a few weeks they probably won't look so bad. The losses only become real if I sell at a low point and lock them in.

MachineGhost often made that claim, and finally some good people were persistent enough that he had to actually address it. He admitted that he couldn't come up with the data. He concluded with something along the lines of "I still think 25% is the max drawdown, but I will no longer claim it to be true". A lot like Dan Rather's "fake but accurate", I think.

I think that fairly well establishes 25% as an upper bound for the PP's max historical drawdown. Whatever the correct number is, it's no more than 25%.

[barrett]

OK, so this is still a bit cursory but from peaktotrough.com I see:

1) A 17.04% drawdown from 1/21/80 to 3/24/80

2) A 9.52% drawdown from 1/21/80 to 3/15/82

Inflation was 12.52% in 1980 and 8.92% in 1981. Source for that is:

http://www.inflation.eu/inflation-rates ... tates.aspx

So, in the first case that 17.04% drawdown was more like 19% in real terms. In the second case that 9.52% drawdown was probably close to 30% in real terms due to very high inflation.

My parameters were 35/15 rebalancing bands with all dividends and interest reinvested. Result View was set to daily.

I believe the reason that Tyler shows the deepest drawdown on portfoliocharts.com as 14% is that he is simplifying things and always using year-end data. The PP did very well both at the beginning of 1980 and again at the end of 1982.

Hope that helps.

[postscript]

Thanks so much for the responses! That's really helpful, and I particularly like the peaktotrough.com site, when I get the time I'm going to dig into the data more.

[Kbg]

I think nominal intraday of 20ish came up
In my backtesting, but it was during the gold meltdown after going nearly vertical the 2 years prior. So, it’s all relative. My planning factor is that a 20%ish DD should be in one’s realm of the possible. However, that’s still not bad compared to many vanilla ports.