The MT version of Habakkuk is plagued with textual issues and corruptions, and is widely regarded as the most mangled text in the Leningrad Codex. For this reason, many turn to more ancient translations such as LXX and Vulgate in order to try to fix the text, while others come up with their own emendations without reference to older translations. Therefore there are wide differences in translations of Habakkuk.
Let's first look at the text, identify the textual issues, and then look at different translations, both popular and academic, describing the reasoning process behind each translation.
אָז חָלַף רוּחַ וַיַּעֲבֹר וְאָשֵׁם זוּ כֹחוֹ לֵאלֹהוֹ׃
ʾāz ḥālap̄ rûaḥ wayyaʿăḇōr wəʾāšēm zû ḵōḥô lēʾlōhô
These 8 words have the following primary senses, with notes quoted verbatim from WBC and McComiskey's commentary  in italics:
ʾāz = then
ḥālap̄ = (verb) fly along, sweep by
rûaḥ = “wind,” “spirit,” or “mind.” Here the problem is that חלף has a masculine form but רוּחַ is feminine. Modern translations interpret ruah as an adverb by emending the text to insert a "like" or ki but this leaves the verb without either a subject or a direct object.
wayyaʿăḇōr = (verb) literally, “and he transgressed" but can also be translated as "and he passed on". From McComiskey:
Since רוּחַ is normally feminine and the
verbal form חָלַף is masculine, the subject of the clause is probably
“he” (the Chaldean), and רוּחַ is to be understood adverbially (“like
the wind”), without being emended to כָּרוּחַ.וַיַּעֲבֹר (and pass on)
should probably be vocalized with the waw-copulative (compare the
LXX’s καὶ διελεύσεται and the Vulgate’s et pertransibit). The
reference is to a change in tactical or strategic planning; compare
the temporary lifting of the siege of Jerusalem in 588 B.C. (Jer.
- wəʾāšēm: (verb) to become guilty. Here is a textual criticism problem. As per the WBC:
ואשם is a 3rd masc. sing. qal pf אשם “and he became guilty.” However
Dead Sea Scroll Habbakuk has וישם G. R. Driver regards both אשם and ישם as probable
dialectical forms of שמם “to be desolate, or appalled, horrified” (cf.
Brownlee, Text, 22–24); NEB follows Driver’s reading.
- zû -- Relative pronoun - this/these. Replacing zû with ze is the second common emendation chosen by some translations, from McComiskey:
It is tempting to read וישם (and he makes) with the Qumran commentary
and replace the Masoretic Text’s relative pronoun זוּ by the
demonstrative זֶה as subject: “And he [this one] makes his strength
his god,” instead of “whose strength is his god.” On זוּ, see GKC
§138g–h. For the suffix ־וֹ (his), as though appended to אֱלוֹהַּ (see
3:3), not אֱלֹהִים (God; plural of majesty), see GKC or 
ḵōḥô - noun with pronomial suffix, "his/their own strength"
lēʾlōhô - noun with pronomial suffix, "his/their own god" preceeding by a l which can be translated as "to, into, as".
Next, let's look how the clauses are grouped together in the Masorah. The question is where to draw the break between the two clauses of this verse. Here is the masorah:
Note that it groups "become guilty" with the first clause, and the second clause should read "this (his/their strength) as his/their power"
וְאָשֵׁם as it stands connotes “and guilty” (as in Gen. 42:21;
2 Sam. 14:13); the Masoretic Text attaches it to what precedes, as
does the Septuagint (καὶ ἐξιλάσεται). But if “guilty” is the sense
here, the word is more naturally construed with what follows (their
guilt consists in worshiping their own power).
Once we understand the various options for emendations, selections of sense, and groupings, we can look at a number of translations and try to understand the choices made:
Then they change course like the wind and pass on—
guilty people, who make their own strength their god!
This translation emends ruah to add a "ki" to get like the wind and it inserts "people" converting the verb "to become guilty" into those who are guilty. It associates guilty to the second clause. It emends zû to ze, and translates it as "who". In the ambiguous choice of the pronomial suffix, the translator selected "they" instead of "he".
Ralph Smith in WBC:
Then he passes by like wind
and passes on.
He became guilty, this one whose
strength is his god.
WBC follows 1. in terms of emendations, but in the ambiguous choice of he or they for the pronomial suffix, WBC elects "he". This translation loses the sense of "changing course".
Then shall his mind change, and he shall pass over, and offend,
[Imputing] this his power unto his god.
(The KJV puts "Imputing" into italics to alert the reader that this was added to help explain the text, while attempting to maintain zû as a demonstrative and prevent it from turning into a subject.)
Here the KJV elects to interpret ruah as mind as in Ezek 22.32, most likely because the connotation is one of changing. It does not emend the text by adding a "ki", but accepts the subj-verb disagreement as this can happen in Hebrew poetry (see GKC, S 144).
Why would the KJV translators conclude this? By consulting the vulgate and LXX which agree on this point: Then shall his spirit be changed, and he shall pass (Vulgate) and LXX: Then the spirit will have a change and pass through. No ancient translation views rûaḥ
as an adverb here -- all accept the s-v disagreement as an artifact of poetry.
Then it translates wayyaʿăḇōr as "and he shall pass over". It follows the MT by grouping wəʾāšēm with the preceding clause as opposed to the following clause, but then translating as "offended" rather than "guilty", which is also one of the sense of this word. Elizabethan English converts a modern "as" with "unto". The (ambiguous) pronomial suffix is interpreted as "he" throughout instead of "they".
Then they sweep like the wind and pass on; they become guilty,
whose might is their god!”
The pronomial suffix is interpret as "they" throughout, rûaḥ interpreted as an adverb. "They become guilty" is placed on the first line, but a sneaky semicolon groups it with the second clause. This translation loses the sense of "changing course".
his spirit changes,
and he becomes angry,
is his god.
The translation maintains the Masorah grouping, with "his" chosen for the pronomial suffix. No adverb is used, so like the KJV, they assume this is a poetic disagreement, but ruah is translated as "spirit changes" which is effectively "mind changes" in terms of the overall sense. However "he becomes angry and sins" is rather far from the text of the single word wəʾāšēm. A clumsy "this" is added in an attempt to add a pronoun but not have it become a subject. One positive aspect of this translation is it maintains the notion of "changes".
Then they sweep by like the wind and go on,
guilty men, whose own might is their god!
The pronomial choice of "he" is made, the grouping follows McComiskey, an emendation to an adverb is chosen, a "men" is added to guilty and the verb is converted into a noun, but a pronomial subject is not added. "sweep by" is the sense translated, which is fine, but the overall translation loses the sense that a change of course is happening, which I believe is important to the interpretation.
“Then they will sweep through like the wind and pass on.
But they will be held guilty,
They whose strength is their god.”
NASB is a little funny by turning the verse into three main clauses instead of two, but still grouping "guilty" with the last group instead of what is changing. It also loses the sense of changing course. "they will be held guilty" is a little far from wəʾāšēm. Wind is taken as an adverb, and the pronomial suffix choice of "they" is maintained throughout.
As we can see, each of these translations follow a certain logic and all of them make some emendations to what is a problematic source text. What is the "true" translation? There is no such thing. But a good translation should try to do the following:
Introduce the idea of changing course. Whether that is specified as the wind changing or the mind changing is less important than the idea of changing course be conveyed, since one is a poetic description for the other.
Convey the idea of passing on in addition to changing course. E.g. I will change course and pass on.
The idea that people who trust in their own strength is the reason why the speaker is changing course and passing by.
The speaker is "he", not "they" -- it is not the Chaldeans that change course, but God.
Then you need to decide whether to follow the Masorah grouping or to follow McComiskey's grouping.
If you follow the Masorah, the idea is roughly as follows:
He [God] will change his mind and keep going (e.g. passing you by) because he is offended [by] those whose strength is their own god.
This would fit into the overall context as follows:
- verses 6-10: God will raise up Chaldeans who are mighty, with powerful horses, that destroy all in their path, etc.
- verse 11: And He [God] will change his mind [to be your savior] and will pass on by, being offended by those [jews] who think their own strength is their God
- verse 12: Then comes a plea from the Prophet "Will you really abandon us? Aren't your promises eternal? We shall not die, You are doing this only for our correction".
In other words, the plea has to be in response to God promising to abandon Israel, not in response to verses describing how powerful the Chaldeans are.
The interpretation that I do not like is ones that use a pronomial suffix of "they", as it suggests it is the Chaldeans changing course and passing by, which of course did not happen. Thus this would make it a false prophecy. The Chaldeans went for Jerusalem and conquered it without diversion or change in military strategy.
Moreover it puts God out of the dialogue, making the response in 1.12 a bit out of place. The whole point of 1.12 is to object to God changing course, because he is everlasting and thus should never change course.
Therefore the KJV is pretty good, although admittedly clumsy in its somewhat maniacal literalism as it manages to faithfully convey the awkwardness of the Hebrew original at the expense of readability. But some of the other translations that insist on smoothing out the verse lose some of the crucial aspects listed above.
1 Image of cantillations from Hab 1.11 Lexham Hebrew Bible: Cantillation Analysis Graphs. Bellingham, WA: Lexham Press, 2016.
 Ralph L. Smith, Micah–Malachi, vol. 32, Word Biblical Commentary (Dallas: Word, Incorporated, 1984), 99.
 O. Palmer Robertson, The Books of Nahum, Habakkuk and Zephaniah, The New International Commentary on the Old Testament (Grand Rapids, MI: Wm. B. Eerdmans Publishing Co., 1990), 150.
 Christo van der Merwe, The Lexham Hebrew-English Interlinear Bible (Bellingham, WA: Lexham Press, 2004), Hab 1:11.
 Ralph L. Smith, Micah–Malachi, vol. 32, Word Biblical Commentary (Dallas: Word, Incorporated, 1984).
 See https://hebrewsyntax.org/hebrew_resources/Beckman%20JC%202013%20%28Subject%20BH%29%20EHLL.pdf
 F. F. Bruce, “Habakkuk,” in The Minor Prophets: An Exegetical and Expository Commentary, ed. Thomas Edward McComiskey (Grand Rapids, MI: Baker Academic, 2009), 846.