Arrow Trader 5
Arrow Trader Core
Arrow Trader eXtreme V2
Arrow Trader eXtreme V1
Arrow Trader Lite 3
before you change any settings here, make sure you understand this page completely!
The slots for the outgoing list directly and heavily affect the way ATX returns traffic to trades.
ATX has a very smart A.I. module. With the help of this module, ATX is able to make an accurate judgement
about the value of a trade. In fact, one part of the A.I. can give a trade a higher or lower value
than it would normally have, in order to try to push a trade to a higher level.
Additionally, by giving trades certain settings, like a hourly-force, a boost, or a forced return trade type
you can force some trades to be above or below other trades in this value list, to give them a higher
or lower priority.
The result is that in the end ATX has a list of trades, sorted on their value.
It will then assign a number of points to (almost) every trade, depending on their position in this value-list.
This is where the outgoing slots come in. You can directly edit how many points (and thus how much % of
the available traffic) a trade should receive. For example, giving slot #1 70% of the points will
result in a massive force to your best trades, but probably won't leave enough traffic for your other trades.
As you can imagine, different settings wil results in very different results.
But using trial & error here is not a smart thing to do since it will most likely kill your site before
you are able to find the best settings. That's why it's important to read this page as it offers some guide-lines.
This table shows the status of the outgoing list for this minute.
Priority trades: These trades receive hits no matter what their quality is like;
For example, a trade with a hourly force or with a high boost.
'Fixed Return' means that this trade receives hits back because the trade type
is set to return x % of hits/clicks
Primary trades: The top 10 (10 being how many slots you have) of your trades, when sorted
on overall value, are your primary trades. ATX will send a surfer to these 10 sites before sending them
to trades of less value, which are the secondary trades. By changing the slots and their values you change
how many trades are in the primary trades list and how much traffic each trade receives.
Secondary trades: All the trades that didn't make it in the 'primary trades' list this minute
are in the secondary trades list. These trades will receive traffic once a surfer has visited all the
trades in the primary list. Once a surfer has visited all your secondary trades as well, the 'exout'
feature kicks in.
Note: Trades that are in the primary or secondary list can also be in the (high)priority list.
Trades that are in the primary list can be in the secondary list the next minut, and vice versa.
Trades that are disabled by the guardian or by the watchlist or by you will not appear in the secondary
With this table you can edit the slots for primary trades. The settings here can have a great effect
on the growth on your site, so here are some tips.
1) To grow your site you need to push your best trades. Action = ReAction. Therefore it's a bad
idea to give every slot the same amount of points (in general). After all, you don't want to send
your 10th best trade the same amount of hits as your first best trade.
A setting like 10 - 10 - 10 - 10 - 10 will keep your site stable, but it doesn't push your trades
to return more.
However, setting 10000 - 100 - 10 - 10 - 10 is also not smart, since that will neglect
almost every trade that isn't good enough to get on that number 1 position.
So the trick is to find the best balance, where you are able to send more than enough to your best
trades, while leaving enough for your 'lesser' trades. ATX comes with rather mild default settings.
I recommend you to increase the amount of points for your top trades. I've had great results whenever
slot #1 was above 30% and slot #2 above 25%.
2) The amount of slots is up to you, but always make sure that you can push your trades.
So adding 100 slots, giving each one 1% isn't going to work.
I do recommend you to have about 2 to 20 (depending on how many trades you have) low-value slots.
Low-value is in this case 1++ point. (bottom 1 point, slot above 1+1, slot above 1+1+1, etc)
That way ATX you can get more control over the trades that would normally fall into your secondary
Here are some examples that will clearify the above
This is a great scheme if you have a small amount of trades that are pretty big.
It will really push them. However, not many sites that are really big only have 5 - 10 trades
so I wouldn't recommend using this.
This scheme will kill almost all your trades, since not many will make it to that top 1 slot.
Of course the trade in position 1 will not really receive 89% of all traffic,
since ATX doesn't send traffic back to where it came from, nor does it send raws.
But around 70 - 80% for spot 1 is not unlikely. Nevertheless, most of your trades will stop sending
because of the poor returns before they get a chance to enjoy the mighty return that spot 1 offers.
This is a nice scheme for a 'normal' site, with 50 - 80 trades.
The push on your top trades is high while enough traffic is left for other trades.
All that's left for me to say now is; experiment!
But be smart. What works for me, might not work for you.
Also, a good scheme still won't make your site grow if your prod is low for example.
And since coming up with a good scheme is one of the harder parts, I recommend you to
make sure all the other parts are dealt with first.
With this table you can add and remove slots to the bottom of your scheme.
Two new features have been added to the outgoing-list settings; bonus mods and custom algorithms.
Both are accessible via the new table with the title extra settings
The first 3 fields (productivity, return and effect) allow you to reward trades that do especially well in a certain area.
The reward will be a higher position in your outgoing list and thus more traffic for those trades.
The bonuses apply to:
1) productivity: the higher the productivity of the trade, the larger the bonus
2) return: the more a trade sends you - compared to how much you've sent him! - the larger the bonus
3) effect: the higher value II rate of a trade, the larger the bonus
For a more detailed explanation see also the optimal settings guide - paragraph 12
ATX can trade on all clicks or just trade clicks. A trade click is a click that was sent to a trade.
These clicks are much more important than gallery clicks, because it are trade clicks that make your site grow.
This number is especially important when you using first / second click to gallery,
because surfers will have to click a lot more to generate trade clicks.
custom return algorithm:
By designing your own return algorithm you take complete control over how the outgoing list gets populated.
You can design a function, and the output of that function is the value that is assigned to a trade.
Then all trades are sorted in descending order, with the highest value on top, and are placed in the outlist
in that order.
The following symbols are supported:
last 60 minutes
rin60 - incoming raw (all) hits
uin60 - incoming unique hits
out60 - outgoing hits to the trade
clk60 - all clicks generated by the trade
cef60 - real / effective clicks generated by the trade
ctt60 - clicks that were sent to a trade
ctg60 - clicks that were sent to a gallery
last 24 hours
rin24 - incoming raw (all) hits
uin24 - incoming unique hits
out24 - outgoing hits to the trade
clk24 - all clicks generated by the trade
cef24 - real / effective clicks generated by the trade
ctt24 - clicks that were sent to a trade
ctg24 - clicks that were sent to a gallery
The following operators are supported:
+ - add 2 numbers
- - substract 2 numbers
* - multiply 2 numbers
/ - divide 2 numbers
() - prioritize a calculation
- All normal math rules apply, so * is calculated before +, etc.
- division by 0 is not possible
- the symbols must be lower case
- don't use spaces
To use a custom algo, set return algorithm to custom function
and copy one of the examples below - or your own function - in the custom function field.
To see what the outcome is of your custom function for every trade, go to menu -> GUI -> edit template.
Then click the 'edit' link for a template and enable the custom formula column.
Give it a orange background and it will stand out nicely. The custom value in the main stats overview is calculated
on the fly so you don't have to wait for the cron to update the stats first.
As you can see in the examples, you can make very simple functions, that just sort on the amount of
to-trade clicks a trade has produced, but you can also make them more complex. Here are a couple of things
to keep in mind while you write your own algo
- Try to include both the 60 minute and 24 hour stats in your function.
When you only work with the 24 hour stats, your algo won't be able to properly handle
trades that start to force a lot of traffic to you. By including the 60 minute stats
you make your algo more responsive. An easy way to make your function stable, but not too stable
is by multiplying the 24 hour stats with a bigger number than the 60 minute stats;
The 60 minute stats are multiplied by 2,
while the 24 hours stats are multiplied by 8.
This makes the 24 hour stats 4 times as important. It's important to find a good balance
between the 60 min stats and 24 hour stats. If you got it wrong, your trades will either be
not stable at all, or too stable and not responsive enough to changes.
- By adding +1 to numbers you're going to divide through you make sure the entire function doesn't return 0
because a division by 0 happened. On big sites this isn't a problem of course.
- The caps like min prod and max return are also enabled when using the 'capped' trade type
- When you're using your own function, the 3 bonus mods are still in effect.
It's recommend that you first set the 3 bonus mods to 100% or it will be a lot harder to verify the outcome of your own functions.
- Since the math-module doesn't support if-then constructions it will be a lot harder to write efficient algorithms,
so don't expect mind-blowing results on the first try.