LindeX market timing

Hey everyone, as anyone with an eye on the LindeX volume knows, there are times of greater and lesser activity.

Anyone other than me choosing strategic $L selling times?

I'm not a daytrader type; in fact I'm pretty stodgy. But it doesn't take a brain surgeon to know when you can move $L 100k near instantly at rate X, or wait days trying to do the same thing at a worse rate than X (put in your favourite exchange rate where the X is).

With US holidays coming, and the beginning of the month coming (pay day for a lot of people) I'm sort of curious what others do.

Will there be someone at LL maintaining a close eye on Supply Linden $L sales, or will they be tucking into turkey while the $L floats? Or is it an automatic process?

Were I to try hard at being sharp with the LindeX, the difference to me is only 20-30 bucks a month in any case, not exactly a big deal and maybe not worth serious effort. But it's something.

Also, for the people maybe planning to buy $L soon - what's the general word about buying some during the US holiday long weekend?

Actually ive often wondered wether there was a certain day or time of the week thats better to sell L$ than others. Up till now i just go for Sunday so i get payed on Friday but id happily change that if there was a better day anyone? (i imagine its the kinda thing people wont want to share though possibly?)

I don't watch it that closely. I sell when I need it, generally.

On a normal non-holiday week (or non crises) best to get your L$ in queue on the sell side by wed night or thurs morning for a fast sale probably friday or saturday. then if you are turning over get into the buy before Sunday for a turnover by tues or wed. You want to be out of the sell side before sunday for sure as Sun-tues are the big days for people selling off thier weekly earnings/stipend. For the same reason its the best time to be in buy.

Holidays can be erratic. If its a holiday and people typically stay home. the volume on the sell side can be increadible and often it seems supply linden is NOT at the wheel. Which means you can get a bit more $ for your L$. If its a holiday where people travel the entire market slows down. Its all a question (particulary sell side) of how many people need to BUY L$. If logins are down or money is tight sell slows. IF you get lucky and there is a huge demand of people wanting to BUY L$ and its a holiday supply linden takes you can however make a killing.

These are all general observations after a couple of years mucking with Lindex. One thing that is certain is that the market and the citizens of SL are NOT completely predictable. I good crises can throw any observed pattern off. Even one good vacation can do it

That's kind of what I've been seeing too, Darkness.

Elanthius had mentioned some... changes in the way Supply Linden had been applied recently, and I've been watching too.

Another point: if there is some major shift regarding the platform, you might notice it first on the LindeX before the news reaches everybody. Kind of an early warning system. Not entirely a perfect link, but if the market suddenly shifts it's probably a good indication to check the news or any other sources one might have.

I don't time it but I use carefully spaced llimit sells to get the best price at a good rate mainly a cent or two below the LEA (Linden enforced average) that way I get at least partial if not total fills on my sells when the price dips but before Supply compensates and by doing that with smaller orders they fill fairly quickly whenever the price drops.

I generally simply sell on the first amount with a quantity higher than 20.000.000L$ or so on the list. I don't mind to wait a few days for a sale to go through. (Esp. as the payout to paypal takes an undefined amount of time anyways.)

I'm not taking the time to constantly watch Lindex for an opportunity to have an extra 1 L$ lower sales point. This comes down to about a 0.3% benefit and is simply not worth my time and energy...

i want to BUY lindens instead of finding inworld ways to earn it, but i will never again unless i can get at least $L275 for my $1US.

will that EVER happen again?

It probably won't since the only reason that the L$ > USD rate isn't 1:1 or less is that they (LL) artifically prop it up since in a free market it would never stay this high both because of the imbalance in sinks vs sources and because the L$ is the fiat currency of all fiat currencies in that like the USD it's a fiat currency but it doesn't have a whole government backing up that a dollar is a doller it just has one company saying that an L$ is an L$.

Full disclosure on this, I am openly pro free marketing it even if that means rebalancing sinks and sources since I only sell L$ and, except for maybe once, never buy it but LL will never do this because A) they make a substantial amount of money selling L$ which they can of course create for free, and B) because even assuming sinks and sources were equalized because of my reasoning on fiat currencies above the natural inclination would be for it to lose (buy) value vs the dollar which would lead to massive inflation and which would make it harder for people to buy L$ to spend in world.

You probably saw this in Zee's feedback thread, but just in case...

He said "...November has returned to normal and to ensure stability we let the L$ appreciate by one L$ and we plan to let it continue to appreciate modestly over time." (/352/14/292421/7.html#post2219415).

And "We did let the L$ appreciate against the USD by one Linden dollar. In general we will seed to let the L$ appreciate a little bit against the USD. We did it at this time to "tighten up" the supply. We recognize the pricing changes are a bit of a shock to the system as is the real world economy. So we're basically giving people a small incentive to hold L$ by letting it appreciate a little." (/352/14/292421/10.html#post2219849)

The best buy and sell rates tend to be stable for several months at a time. So if you place limit orders at those rates, you will - with rare exception - get the best possible rate within a week to ten days. The only thing that changes is how quickly those orders are filled. However, given that there's no real opportunity to store funds in an interest bearing account until you place an order, that time delay is only relevant if you have some urgent need. I don't really see how any sort of market timing strategy can apply here.

However, if what you're doing is trading back and forth regularly to make a gain on the spread, then there's an interesting question as to whether going one L$/USD worse will pay off. Say, for example, you're trading quantities that earn you about 0.1% profit per round trip. If you average a week per round trip, you're making 5.2%/year (ignoring compounding and the fractional week). If you could reduce the round trip time to 6 days from 7, you'd be making almost 6.1%, assuming the same 0.1% per trip. But if you achieve that reduction in time by using a 1L$/USD reduction in profit, that's very roughly a reduction from 0.1% to 0.096%. If I've done my quickie calculations correctly, that changes the annual yield to about 5.86% - still an improvement over the 5.2%.

The timing enters the picture by asking when should you give up that 1L$/USD? On the buy side, sell side, or both? If you could quantify an effective timing strategy for this question, you could squeeze out more profit.

There's a simpler approach. Go to the Market Data page, look at the Min Rate column (for selling), and use that value. That's the best selling rate. It might take longer for the sale to go through, but it gets you that 0.3% benefit without the time and energy expenditure.

That's 99/100 times basically the same value as the top 'bump' in the quantities I meant. With 'watching', I meant trying to follow fluctuations in the Lindex to predict the best day to sell (where min rate is lowest).

This thread inspired me to create this: http://ninjaland.net/lindex-market-data/

Not sure how reliable it will be or what I'm going to do when there are 20,000 rows but until then it should updated hourly.

well in my case it only amounts to pennies anyway, so i dont worry about it much.

A market shift USED to be a very good indicator. It is much less so now. I remember in the early days of the 2-sided lindex massive shifts of 30 points or more at the slightest hint of something. Now between supply linden and the sheer volume of the lindex the shifts are so small and short lived they tend to not be terribly significant. Even at the height of the OS mess and the ensuing selloff. the shift was only 1-2 points and the market recovered within a few days. Even then it was pretty much sell side only with people jokeying to get thier assetts sold off and money out of SL. Buy side seems to be continuing players or Lindex flippers who want the most L$ for this US$ and doesnt move much unless someone gets impatient.

It takes a massive number of people acting in concert or a massive influx/withholding of cash by supply linden to effect a permanent shift in the lindex these days. That said we did have a 1 point shift recently. WHo engineered it is up for debate. The lindens claim they did. I tend to suspect it started with residents and by the time the lindens noticed it was well underway and they decided to roll with it. In any case they now claim it was thier doing. Might have been. We the residents have no way of knowing.

/me still remembers King Phil insisting that 250L/1 US$ was the ideal

I've never even thought about timing as a factor on the Lindex. The only variable that I consider is how quickly I want/need some US$ == what rate to set my limit sell at.

I typically set it at the rate where the current backlog is at, and it goes through in a day or two. If I'm in a hurry, I'll sell for L$1 above wherever the current backlog is at to get it within an hour. However if I don't need it for a week or more and the backlog seems fairly divided, I'll put it at the lower rate and wait a bit longer.

For the amounts that I work with, an exchange difference of L$1 would result in a difference of no more than US$1 or $2 per month either way, so I consider it to be fairly negligible.

-Atashi

That is actually a pretty good way for most people to approach it.

When I was "day trading" how fast I could rollover a sell to a buy and what the point spread was between sell price and buy price were everything. So I payed very close attention to it. Close enough to notice trends anyways. With a more stable lindex we got a much smaller point spread which pretty much locks many players out of day trading for profit as the amounts of money involved have got huge before you see even a small profit. But when I was doing it you could make a nice profit with even just a 50k L$ flip. Now you need 100k to break even roughly on the typical 11 point spread we have. and MUCH MUCH more to make enough profit to make it worth worrying about doing. In other words not worth it.

The change of 1L$/USD corresponds to $4 per 1000 USD. I usually don't worry about an extra L$1/USD. However, if your operating costs are high (such as if you are the owner of a particular large Victorian Steampunk estate) I would imagine it would make a difference of hundreds of dollars a year.

If it helps, in my sales I find that sales peak of course on the weekend, but also a smaller spike on late Wednesday and early Thursday. For the weekend, interestingly it used to be big on Saturday and with smaller increases on Friday and Sunday, but interestingly it seems that the weekend peak has shifted to late Saturday and early Sunday. More people drinking on Friday night perhaps?

If it is given that L$ demand is correlated to sales volume, I'd expect those periods to be the best to sell L$.

Of course, Linden makes an absolute killing when the economy grows and they can print new Lindens which are almost entirely profit. I would like to see an eventual VW currency based on something such as gold with a reserve ratio of 1:1 (every VWGMG (virtual world Gold Milligram backed by a RL weight of the same Gold). The L$ is a double jeopardy, a fiat currency reliant on another fiat currency (USD usually, as you can't buy food or pay taxed with L$.)

Your tuesday/wed sales increase may well be due in part to stipends going out on tuesdays.

And yep the weekend spike does correlate to fast L$ sales on lindex.

day trading as I described it is RISKY. Very risky these days. Although the point spread remains pretty constant reducing that part of the risk. The fact is you have to keep several THOUSAND US$ on your account to make a profit. Not Particularly safe since LL can take away anyones account whenever they like. Also given the hacking and such that HAS happened in SL, I am not comfy with large balances sitting on my account to be swiped. Hence it isnt something I am willing to risk anymore.

Given that the best price for the L$ varies about 1L$ a month, I just put my L$ for sale at the day's best price and if it takes a few days or a week to sell, I don't care.

Zee was kinda informative regarding the $L drift thing, and that's very interesting. It *does* look like there's a man behind the curtain sometimes; or maybe I have just been staring at the numbers too long and getting dizzy. Sort of like watching the static on an old TV...

Also fascinating how the velocity of money helps day trading percentages, though I'd never get near that. Last thing I need to do is go 'oops' on some crazy day-trade and have 44 regions go dark. That's like end-of-the-world-scenario stuff, no thanks! The money I lose not doing that is probably what the stress medication would cost... besides, I'm too thickheaded to fully comprehend what the data is saying even if it's right smack in front of me. Elanthius, is there some formula or something that would drag out the velocity of money on LindeX or the current percentage per $L 1M arbitrage or something? Heh I suppose selling $L 1 at midnight and timing when the $L sale email comes in might do it, but what a pain (not to mention the 30 cent fee or whatever it is that ruins that idea).

And yeah I have been looking at the math alright. It's not just tier it's cashflow. Due to openspace price changes and some other estate business issues I'll be looking at an $L cashflow in the neighbourhood of 12,500 USD/month in January. 4 per thousand is what, 50 bucks difference. Or 600 bucks a year.

Compare that to the 3.5% currency conversion fees, which are 437.50 a month on 12,500 worth of $L. Multiply by 12, that's 5,250 USD annual in financial fees just from my estate alone, which actually isn't all that big compared to a lot of them out there. Not a bad haul for an automated system...

* * * * *

I wonder what that 3.5% really makes overall off Lindex. Let's do a really dumb calculation.

Presume an average of $L 70 million a day, which is $L 2.1 billion a month or $L 25 billion a year, roughly. That's a guess, you can look at the LindeX market data and kinda guestimate what average volume is yourself. 3.5% of $L 25 billion is $L 875 million. Or maybe about 3.3 million very real dollars? Whoa. Did I mess up the math anywhere? My RL biz processes credit cards also; yeah there are some pretty hefty expenses, but, still! Seems they got the salary for the guy who does the paypal stuff covered...




Some days I'd give a whooole bunch of $L for a pizza to suddenly appear.

Thats probably a pretty accurate estimate on the sale side for a busy saturday night. (based on my own observations of how fast new money gets listed and how fast it takes a listing to clear). Might actually be a bit low.

One thing not to discount in this though is LL makes a fee on the seller selling their L$ but they ALSO make a fee on the person buying them ($.30 per transaction). Just think of all those people placing one 2-5 US$ order in buy after another and the scale of LL's Lindex earnings gets HUGE.

Sell L$1million at 262 with a 3.5% fee, and you'll get US$3683.20.
Buy L$ at 273 with US$3682.20 and a 30 cent fee, and you'll get L$1,005,431.
That's a 0.5432% profit.

Sounds like it's not worth doing, but lets look further. When Supply Linden is doing his thing and keeping the people on the buy side happy, you can turn that around a couple times a week. Compound that interest 100 times in a year, and you get 71.88% annual profit. Occasionally you'll miss a couple weeks while Supply Linden takes a nap. Sometimes the market will shift by L$1 and you'll lose the value of your last flip, but even if you only average 1 flip per week the annual profit will be 31.1%. Compare that to your RL portfolio.

This is why there's always a ton of people on the limit buy side waiting for Supply Linden to do a huge market sell.

I'm almost certain Supply Linden does not do market sells. As for your other stuff, it's all true and I wrote a little program that does it automatically for me for any money I happen to have spare. Usually takes about a week for a full round trip though.

I've been trying to work up the energy to create a program where lots of people can donate tiny amounts of money to me and I'll process it through the LindeX and pay them back the principle and the 5% profit after each round trip. Probably would get into trouble for being too bank like though.