Searching for a Accumulation/Distribution Buying Pressure script

[Tamza]

Does someone perhaps have a study for NT8 that is similar to the AccumDistBuyPr study in TD Ameritrade's ThinkorSwim? I've searched the NT Ecosystem, but am unable to find something similar. This is the best definition I can find to explain it (taken from ThinkorSwim's Learning Center):

• The indicator is in Accumulation mode when the Close priсe of the current bar is greater than that of the previous bar. Accumulation is calculated as the difference between the current Close and lowest of the two values: current Low and previous Close.
• The indicator is in Distribution mode when the Close priсe of the current bar is lower than that of the previous bar. Distribution is calculated as the difference between the current Close and greatest of the two values: current High and previous Close.
• The final value is equal to the algebraic sum of all prior Accumulation and Distribution values. Bars whose Close price is equal to that of the previous bar add zero the the final result.

Thanks to anyone for their assistance!

[NinjaTrader_ChrisL]

Hello Tamza, thansk for your post.

I am not able to find anything similar, regrettably. This post will remain open for others to give input.

Kind regards.

[Hyper]

Hi. I have a lot of experience with Inventory Analysis, which includes
"Accumulation and Distribution" i.e. Market Maker Long/Short analysis,
including whether Market Maker's inventory over a defined period
of time, is Making or Losing money, momentarily, against its Cost Basis.

The "Easy, and not so Accurate" approach to this is simply to Timestamp
"signed volume" so transactions to the BID are Retail Sellers, and MM
(Market Maker) buys here, while transactions to the ASK are Retail Buyers
and MM sells to them. Each such event should be timestamped, and
retained in a "moving window" or list, usually in memory, so that you
could take the easy way out; and simply for ALL transactions between
time X and time Y, just sum up the Volumes. I always do that from MM's
perspective, so that a Buy from a Retail Seller, moves MM in the "Long"
direction (say, a positive number); and a Sell from a Retail Buyer contributes
negative volume in the "Short" direction.

In this "easy version" of the algorithm; you have an estimate of the Net
Accumulation or Distribution of "Time and Sales" so that you can see
whether MM is "nearly flat", Long (on the positive side) or Short (on the
negative side) for Net Volume of "Inventory".

HOWEVER, 1) that is very inaccurate, and 2) while knowing that MM is "Long" or
"Short" over a period of time, seems interesting and useful; in actual practice,
it is not. So these are 2 shortcomings in this "easy version" of estimate the
Net Accumulation or Distribution of contract volume, from MM's perspective.

You might think that if MM had bought loads of contracts from Retail Sellers,
and is now "Long" in inventory (and wanting to sell that inventory off at a
higher price) that it would be a "proxy" for the concept of Price Support,
so now you are going to "know" when to BUY in this "Long Support region",
but, think again, and you'll see this is not very useful information; even if
it is accurate.

Consider your own trading. You might be Long 3 contracts, in some symbol,
and so the question arises; are you "happy" or are you "sad"?? That completely
depends upon whether you are losing money, or making money on your
"inventory". If you're losing money; it's a problem. But if you're in profit,
then you're nice and relaxed.

Market Maker is the same, but on a completely different order of magnitude of
scale. If Market Maker is holding 1000 contracts "Long" against the Retail Market,
then surely She won't drop the price that much further; and that will represent
some estimate of "Support".

But ONLY if MM is BELOW Her Cost Basis, is MM losing money. When that situation
occurs, then you can begin to understand that MM is not only holding Long inventory
in this example; but is actually Losing money or what I call "in Long Risk" in this
scenario.

But in order to know that, you have to 1) more accurately calculate Net Inventory,
so that 2) you can arrive at an estimate of MM's "Cost Basis" or Volume Weighted
Average Long Buying Price, so that you can then meaningfully say, Yes, MM is
not only Long, but is Losing money since Market Price is below MM's Cost Basis,
over the period of time that Inventory is being evaluated.

I'm not gonna go too much further into this, only to say that you'll need to separate
Inventory into all of the MM Buy transactions, and MM Sell transactions. In the
case of Long inventory (you have to firstly decide whether the aggregate is
either Long or Short, in order to do this correctly). So you evaluate it as Long
Inventory. In that case, I would take the Highest Buy transaction and see
whether it could be matched with a Sell transaction volume which is at a Higher
Price. If so, then that Volume is Removed from your Inventory, since it is
profit taken by MM.

Without too many further details, by doing this from highest to lowest price,
then you have a Much more Accurate idea of True Net Inventory; and you are
left with ONLY Buy transactions, on which you can calculate a Cost Basis.
That then allows you to say whether MM is Long, yes, but whether that Long
Inventory is in profit; or whether it is in "Long Risk" and might realistically
represent Support.

More than that, I'm sure you don't want to know. So just buying an Accum/Dist
indicator is very unlikely to give you anything you can use in trading; unless
you take into considerations the concepts I discussed above.

Best of Luck, hyperscalper

[Hyper]

oops, deleted, posted in wrong thread

[Tamza]

Thanks,Chris, I appreciate you help.

Hyper, I'm grateful for you taking the time to share your knowledge about this. I see that there is a lot more to accumulation/distribution than I realized; I'll look further into it, at least to get a better understanding of how market making works, etc. Thank you for this!

[Hyper]

Tamza, I wish you well; any questions, or if you think you want to code up
something like this, feel free to ask; but let me say one thing:

Knowing the Inventory Net Imbalance; and even knowing the
Risk Levels (aka "losing money on Inventory") which Market Maker
may be experiencing is only what I call "situational awareness".

Yes, they are in Long Risk, and it may amount to a million Bucks;
and that's a Support indication; but knowing exactly how "deep"
into Risk they may be willing to dive, before raising the price, is
fairly imprecise.

Other factors, such as Depth of Market analytics, or even "Big Lot"
large trades directional detection, are more helpful in pinpointing
exactly when that market is going to start Lifting.

Everybody knows there is high volatility at both major tops and bottoms,
and that is associated with faster Accumulation/Distribution, and that
is where Market Maker will put the most size on the line; to get what
you can think of as "one last gulp of volume" before the Price Trend
is reversed.

So any time you can get some indications "other than just Price", such
as Volume related behavior, e.g. Accumulation rates or Distribution rates,
or size bias pattern changes on the DOM, etc. then you may be able
to gain an edge. Let's face it: Everybody on the planet sees Price, and
so clearly Price movements are used to confuse traders; and something
"deeper" (which is not Price) is needed to confirm a real top or a bottom.

[edit] Pure "Price Action Traders" may tell you that "Patterns" of Price
movements may be predictive (e.g. double tops, etc) and I am willing
to concede that is likely to be true. But since Price is what nearly all
Indicators are driven by; and what almost all Traders use; it's worth
considering that "deeper, non-price based indicators" are what you
should rightly be looking for ! So maybe you can find something
which fits the bill... Best of luck ! Like many traders, I've spent
a decade or two; and finding that "magical indicator" is always
a constant effort.

hyperscalper